Skip to main content

Full text of "The esthetic basis of Greek art of the fifth and fourth centuries B.C"

See other formats








Professor of Classical Archaeology 
in Bryn Mawr College 

So. 10-3 

Bryn Mawr, Pennsylvania 

New York, Bombay, Calcutta, and Madra? 






ART ' . 1 

TION 28 

TURE . . 76 











The present monograph is a critique not 

of artistic taste but of artistic behaviour. 

It makes no attempt to eulogise or appre- 

ciate or evaluate ancient Greek art, but 

solely to examine Greek artistic procedure 

and by such an examination to arrive at 

some fundamental esthetic problems and 


Such a study, since it tries to be funda- 

mental, must aim at a method suffi- 

ciently general to be applicable not to 

Greek sculpture only, but to sculpture in 

all times, and not to Greek architecture 

alone, but to architecture the world over. 

The chapters which deal with these arts 

overpass, therefore, the boundaries of 

Hellenic antiquities and attempt an esthetic 

critique of sculpture and architecture as 

human (and not merely as Grecian) artis- 

tic activities. But the starting-point for 

theorising has always been Greek practice. 





Those who are familiar with the little 

of value which ancient esthetic speculation 
offers, will perhaps have noticed that the 
best of this speculation comes, not from 
the philosophers, but from the practicing 
artists themselves and from the historians 

of art. I have tried to follow and incor- 

porate every such hint or indication of the 
intellectual attitude of the ancient artists 

toward their craft. 

The bibliography of my subject is very 
nearly negligible. One debt (outside of 
Greek archaeology altogether) is, however, 
a heavy one; and I wish to acknowledge 
great obligation to the keen and serious 
dialectic which distinguishes Geoffrey 
Scott's Architecture of Humanism. 

R. C. 

Bryn Mawr College 
September, 1921 








SPECIMENS of Greek hand-mirrors may be 

An intro- 

found in any of the great museums of 

ductory in- 

classical art. The worn and patina-cov- 

ered disk was once a plate of burnished 

bronze. To it was fastened a handle, often 

terminating in a base on which the mirror 

could be made to stand upright. A circle 

to reflect the circle of the human face, a 

handle by which to grasp and turn the 

disk, a support on which to stand it 

these three elements arose out of the ser- 

vice for which mirrors are made. With the 

Greeks, art was closely wedded to mere 

utility; and the modern museum-goer can 

discover something of the essential beha- 

viour of Greek art if he pauses to see what 

the ancient artist contributed to these 

three utilitarian elements of an old-time 





Unity of 

Jnity of 

Disk, handle, and base a circle, a shaft, 
and a spreading bottom joined and made 
one, since all are parts of the same mirror. 
Joined, but how? By rivets, usually. 
But these are mere material links, holding 
disk to handle and handle to base. But 
low join a circle to a columnar shaft and 
this in turn to a spreading stand? It is 
easy to fuse the matter; but how shall we 
fuse the form? The straight boundary- 
lines of the shaft should swing out and 
curve and in their curve go over into the 
disk's full circle; the broad bottom of the 
stand should rise up and narrow to the 
shaft's straight ascent. By some sort of 
geometric transition each characteristic 
shape must pass over into the adjoining 
one, and a continuity of outline must hold 
all three forms together into one. 

This is simple and good, so far; but is 
there no more to do? are the surfaces 
within these interflowing boundary-lines to 
be left unformed? If the shaft holds up the 
disk, it can make that service compre- 
hensible to us not merely by the patent 



farts of gravitational support, but in a 
more direct visual manner. 

Perhaps the most effectual way to sug- 
gest to the beholder this function of sup- 
port would be to model the shaft into the 
likeness of a human being who carries the 
disk upon his head. Looking at such a 
representation, one could hardly fail to 
appreciate the relation between shaft and 
disk. We might claim that gravitation had 
been as it were visualised and given to us 
directly by a sympathetic analogy. An- 
other way to suggest this function of sup- 
port might be to seek an analogy with 
architecture and to carve the shaft like a 
fluted column. Our apprehension would 
then be less immediate, perhaps, since the 
appeal to human experience is no longer 
direct; but there might be a compensation 
in the up-and-down of the fluting-channels 
and the emphatic simple shape divested of 
all irrelevant associations. Best of all, 
these two methods might be combined. 
The shaft could be modelled as a human 
figure architecturally formalised by means 
of vertical ridges and channelings and a 




general column-like outline. Then the ap- 
peal would be two-fold: a linear unpic- 
torial suggestion borrowed from architec- 
ture would be fused with an animate rep- 
resentation, and both would be perfectly 
consonant with the simple mechanical func- 
tion of a mirror-stand. 

There would remain only the transition 
from the shaft to the disk, to be "ani- 
mised" and presented in some sort of pic- 
torial analogy. On an ancient mirror 
loaned to the* Metropolitan Museum in 
New York two little winged love-gods fly 
above either shoulder of the central figure 
and thereby carry the outline over into the 
curve of the disk, spreading the rectilinear 
shaft out into a circle. To justify the 
occurrence of these Erotes, the human- 
figured shaft has become Aphrodite, whose 
presence upon a mirror-support needs no 
excuse. So, everything has been taken 
into account structure and shape and 
meaning and use and the result is not a 
mere blind embellishment to make pretti- 
ness out of plainness, but an ordered, con- 
sonant, and intellectual humanisation of a 



number of abstract and not exactly ob- 
vious properties and relations. 

And therein, it would seem, lies much of 
the characteristic behavior of Greek art 
in rethinking certain essential matters of 
structure, purpose, and fitness, and in re- 
embodying them in a fusion of geometric 
form with pictorial illusion. 

"A charming thing!" comments the 
museum-goer who pauses to look at the old 
mirror. But it is not really a question of 
charm; it is a question of a language which 
speaks through the eye instead of through 
the ear, but appeals just as insistently and 
directly to the intelligence and the emotions 
of the stirred imagination as do the spoken 
syllables of intelligible speech. As a prod- 
uct of artistic craftsmanship, it was made 
not for a vague delight, but for sharp and 
definite emotional comprehension. It is 
not such stuff as dreams are made on, but 
a product of logical thought of a very par- 
ticular bent, inventing a visual embodi- 
ment for itself (in conformity, be it added, 
with a sense for loveliness). 

Into the inner workings of this " sense 

Greek art 


E S T H E T I C B A S I S 

and open to 

for loveliness" it is difficult to penetrate. 
Esthetic philosophy has wasted its energy 
in rather a priori and abstract discussions 
of the nature of beauty. It has perhaps 
succeeded in defining its realm, but it has 
scarcely managed to reduce it to anything 
other than itself. Beyond the restrictions 
of such theorising, the canon of taste con- 
tinues to operate according to its own good 

But if we are determined not to enter 
upon discussions on the nature of the beau- 
tiful in art, but confine ourselves to exam- 
ining the working of this " logical thought 
of a very particular bent, inventing a vis- 
ual embodiment for itself," we shall be 
dealing with a more intellectual process 
amenable to intellectual analysis, about 
which words can be written and read and 
definite ideas formulated. We shall find 
that we are dealing with something very 
concrete and very tangible, concerning 
which statements may be made with a 
content demonstrably true or false. By 
means of such an inquiry we shall not be 
able to answer the question, "What con- 




stitutes good art?" but we should be able 
to reply intelligently and definitely to the 
question, "What does the artistic process 
do? how does it behave?" In the specific 
field of ancient Greek art, this is the ques- 
ion to which this monograph attempts an 

which is the 
purpose of 
this book 




A more 

Art is the 
who makes 
life of the 


The impulse which drives man to employ 
imitative art for the purpose which we call 
decoration apart from those utilitarian 
ends which magic or superstition may 
suggest is (apparently) the desire to ani- 
mate the inanimate and fill the unliving 
world with living forms. It would seem 
that this might be only another manifesta- 
tion of that tendency to regard everything 
as an animate agency which is at the back 
of pagan religious superstition. Just as 
the primitive man peoples his stones and 
trees, his hills and rivers, with living spirits 
having always more or -less human forms 
and passions, until the physical world of 
unfeeling objects about him lives for him 
as a more intelligible (even though not al- 
ways more companionable) place of hu- 
rnanised demons, so the craftsman and 
artist seizes upon inert wood and stone and 
metal, and converts it into the likeness of 
animate things, that he, too, may people 
his surroundings and raise the lifeless to 
guise of the living. To see a spirit in every 



tree and rock and fountain-head and cave 
and stream, and to see a resemblance to a 
living creature in every chance outline and 
surface and pattern these are scarcely one 
and the same. But it may be that both 
hark back ultimately to the same primitive 
instinct. The child makes persons out of 
the senseless furnishings of his nursery, as 
the savage can make persons out of his 
environing objects and, with fear and fail- 
ure working on him, can beget him a fear- 
ful theogony. But where the savage takes 
his animism for terrific and malevolent 
truth and believes in all the spirits of wind 
and corn, there is always in art this reas- 
suring measure of sophistication, that 
craftsman and public know that what is 
imitated is not truly real, but a work of un- 
tyrannised fancy. 1 The artist, working for 
delight and not from fear, gives us again 
the child's pleasure of peopling our dull 
surroundings with interesting and kindly 
life; and with this simple pleasure we fuse 
those more intelligent and thoughtful de- 
lights that come with artistic contempla- 





This conversion of things whose shape is 

but arbitrary and conventional into the il- 

lusion of nature-given shapes of living 

things plays upon us with a thrill that I can 

best illustrate by offering 

an instance of 

the converse process of 

converting the 

chance outlines of living 

things into the 

stately, immovable, and unbetterable con- 

tours of a shape of art. 

A poet, seeing 

girls with water-jugs on their heads, has 

written : 

Voici bien, O Jacob, le 

geste dont tes 



Savent, en avancant d 

'un pas jamais 

trop prompt, 

Soutenir noblement 1'amphore sur leur 


Elles vont, avec un sourire taciturne, 

Et leur forme s'ajoute 

d la forme de 


Et tout leur corps n'est 

plus qu'un vase 

svelte, auquel 

Le bras leve dessine une anse sur le del! 

Much of the wizardry 

of poetry rests 

upon this same instinct for animising the 

inanimate by discovering 

analogies with 





living things or informing formless -things 
with familiar shapes : 

He will watch from dawn to gloom 
The lake-reflected sun illume 
The yellow bees in the ivy-bloom, 
Nor heed nor see what things they be; 
But from these create he can 
Forms more real than living man, 
Nurslings of immortality ! 

One might almost maintain that it is the 
poet's office (like the artist's) to animise 
the dead world, his medium being, instead 
of direct visual presentation, the verbal 
suggestion attaching to epithet, analogy, 
and metaphor. The truth of such a state- 
ment would of course have to be proved by 
an appeal to the whole corpus poeticum; 
but it may be briefly substantiated by 
quoting a strophe of Shelley's Ode to the 
West Wind as an outstanding instance of 
this envitalising instinct of poetic art : 

O wild West Wind, thou breath of 

Autumn's being, 
Thou, from whose unseen presence the 

leaves dead 




Are driven, like ghosts from an enchanter 


Yellow, and black, and pale, and hectic 


Pestilence-stricken multitudes: O thou, 

Who chariotest to their dark wintry bed 

The winged seeds, where they lie cold and 


Each like a corpse within its grave, until 

Thine azure sister of the Spring shall 


Her clarion o'er the dreaming earth, and 


(Driving sweet buds like flocks to feed in 


With living hues and odours plain and 


Wild Spirit, which art moving every- 


Destroyer and preserver; hear, oh, hear! 

Every element of the windy autumn land- 

scape turns to living forrn and animate 

activity. Nor does the rest of the poem 

appreciably abate this same process of 

verbal sorcery. 

Craftsmen in all times and places have 

pursued this practice of artistic animism. 





whose chief procedure is the discovery of 

analogies between the form of the objects 

to be decorated and that of some denizen 

of the animate world. Just as the poet 

clutches at every suggestion of similarity 

through which to create metaphors by 

analogy with living things, so the Greek 

decorative craftsman sought analogies be- 

tween the shapes of the inanimate objects 

on which he worked and those of animate 

beings into whose likeness he could modify 

them. The "legs" of chairs become the 

legs and claws of beasts; the stand of a 

mirror becomes a human being; the shaft 

of a column becomes a giant or a maiden 

with a basket on her head; a bracelet be- 

comes a twisted snake; the "nose" of a 

lamp becomes the muzzle of a beast; the 

handles of a cauldron become two wrestlers 

leaning over at grips with each other. In 

other epochs, weather-vanes become cock 

or fish; salt-cellars are ships or windmills 

(and therefore half-way animate, since they 

now can appear to move); nut-crackers 

are dwarves; andirons are beasts; door- 

knockers are demons; bellows are dragons; 





and so on, without limit, but by no means 
without reason. 

Just so in Egyptian art a spoon becomes 
a girl lying prone with outstretched arms 
holding a duck whose movable wings are 
covers for the bowl of the spoon. In 
mediaeval illuminations the initial letters 
of manuscripts suggested grotesque animal 
analogies to the embellishing copyist. 

In the Scythian art of the Russian 
steppes this animistic principle finds a 
characteristic and strange complication: 
the various parts of an animate representa- 
tion are in their turn subjected to a similar 
scrutiny for possible analogies, as when the 
antlers of a deer become the long-necked 
heads of birds and the muscle-folds at the 
shoulders acquire beaks and eyes, or the 
wings of winged lions are decorated to the 
semblance of fish. In the Maya art of 
Central America the very same practice 
obtains; each portion of a pictured person 
or object becomes a new field on which the 
artist can uncurb his imagination until the 
whole surface of a design is a puzzle-pic- 



ture of ramifications into avian, reptilian, 
floral, and human motives. 

Greek art did not permit itself these 
mixed metaphors until a late period. 
There was a sound esthetic practice which 
refrained from treating an animate repre- 
sentation as a field for further invention, 
since to do so would be to destroy the or- 
ganic unity and animate value of the orig- 
inal design. The esthetic objection to 
Scythian and Mayan decoration is that it 
fails in the primary purpose of raising in- 
animate objects to an animate status. We 
can see how a quiver-cover might become 
deer or gryphon if we are ready .to allow 
that inanimate matter may be informed to 
an organic structure; but we cannot under- 
stand how a wing can be a fish or a fish be 
a wing, because each has animate existence 
in its own right. For the same reason an 
animate representation cannot carry irrel- 
evant decoration. Greek taste would not 
have countenanced the Vettersfelde fish, 
whose scaly body is adorned with animal 
motives in good early Ionic style, while 
his double tail curls up into rams' heads; 




nor would it have approved the Kul Oba 
gold deer, whose flanks are overlaid with 
image of gryphon, hare, lion, and dog. Yet 
both these objects were probably made by 
Greeks for barbarian trade. 

In the decorative arts, as we have seen 
the representational subject-matter is sug- 
gested by the chance shape and appearance 
and purpose of the material to be decorated. 
In the "pure" arts, such as sculpture and 
painting when they are producing works 
which have no other use or purpose than 
their purely artistic and esthetic aim, the 
choice of subject-matter is not dependent 
on the invention of animate analogies. 
Yet theirs is the same underlying impulse 
of shaping the inert and merely material 
into the illusion of animate existence. I 
cannot therefore see much merit in the 
current distinction between "pure" and 
"applied" arts, between the work of art 
which exists in its own right merely for the 
artistic pleasures that it gives and the 
work made for a practical and utilitarian 
purpose but embellished with such artistic 
touches as may be conveniently and in- 



cidentally applied. This is the current 
conception in the popular mind; but of 
this distinction there comes but little good. 
Sooner or later it leads to the impression 
that art is something external to material 
objects, something which may be added or 
omitted at will, like the interior furnishings 
of a house, and so occasions the idea of the 
Superfluousness of Things Esthetic an 
idea which is only too commonly the pop- 
ular attitude toward art nowadays. This 
idea still has a tenacious hold upon archi- 
tecture and even upon professional archi- 
tects, who are prone to act as though they 
thought that artistic effects were something 
accessory which were only added for the 
sake of good appearances. 

In "pure" and " applied" art alike, we 
shall find among the Greeks this same 
process of animism,, so that it should occa- 
sion little surprise that the subject-matter 
of Greek art is almost wholly confined to 
the world of men and animals and that the 
Greek artists showed little fondness for 
portraying inanimate nature. So strongly 
did they feel their province to be with the 


art imitate 
the ani- 
Greek art 
most of all 




higher animate forms that they rethought 
the inanimate, wherever possible, into ani- 
mate terms which could be substituted for 
it. This practice, which may be consid- 
ered fundamental in their attitude toward 
art, necessitated a sort of symbolism in 
terms of animate objects. Thus, though a 
chevron or scroll may often typify water, 
the favorite method was to select some 
aquatic animal. The beast stands for its 
habitat. A crab at the foot of a boulder 
gives us the Saronic Gulf washing the base 
of the Scironian cliffs; dolphins stand for 
the sea, freshwater fish for rivers, a heron 
for marshland, a swan for a lake. At times 
the symbolism seems forced, as when a 
maid stands for a spring of virgin water, 
or a man-headed bull for a roaring stream; 
but that is because we to-day have dropped 
out an essential link in the chain and for- 
gotten how the Greek mind peopled foun- 
tain, tree, and hill, and river with local 
divinities an instance of this same proc- 
ess of animising natural phenomena by 
visualising them in animal or human form. 
Largely because of this process (but also 



because the Greek was little interested in 
showing in his art those objects whose 
typical form he could not clearly catch) 
Greek artists paid so little attention to 
landscape. A large part of the delight of 
landscape-composition lies in the synthesis 
of intelligible but rather haphazard ap- 
pearances into a design full of the illusion 
of spatial depth and into a color harmony 
full of the suggestion of lights and shadows ; 
but of typical form there is little, and of 
human analogy there is none. A Greek 
mountain has only its skyline; the rest of 
it is a flat plane with haphazard splotches 
of rock, herbage, and tree. Each tree may 
have a characteristic shape when examined 
singly and close at hand; but at a dis- 
tance, merged with others, it contributes 
vaguely to a whole which has no unity of 
structure or necessity of interior form. 
On the Arezzo amphora where Pelops car- 
ries Hippodameia on his chariot, the es- 
sential constituents of the landscape are 
timidly and disjointedly given; above, a 
lightly scratched profile typifies mountains 
against the sky (those marvellous moun- 





tains of the opposing shoreland that 
stand so glorious across the blue Corinthian 
Gulf); by the chariot two isolated tree- 
forms suggest the olive-groves of the 
Achaean shoreland; in front, a wave pat- 
tern and a plunging dolphin symbolise the 
windy blue reach of sea. But there is no 
synthesis of a seen and solid landscape. 
The actual and complex scenic setting was 
apparently formless to the Greek mind: 
there was no way of drawing it, and there 
was no way of animising it. 

It may be remarked in passing that our 
discussion suggests that a scrutiny of Greek 
drawings and sculpture will throw no light 
on that often-raised and rather fruitless 
question whether the ancients were sen- 
sitive to the beauties of Nature. From the 
striking absence of landscape scenes, 
whether of forest or cornland or olive groves 
or rocky shores or pine-covered heights, no 
conclusion may be drawn as to the senti- 
ments of the ancient people who lived in 
that (still to-day) marvellously beautiful 
country-side. 2 

In Greek art Nature is rethought in 



terms of human or animal analogy. Greek 
coins are the locus classicus for a study of 
the workings of this process. 

We of to-day are generally content to 
convey our meaning on our coins by en- 
graving verbal information: "United 
States of America One Dime E Pluribus 
Unum Liberty 1921 In God We Trust." 
To this explicit announcement the "art" 
is fairly superfluous, something by way of 
ornament and cultural tradition. Any 
more or less frigid allegory will serve; a 
young woman with a Phrygian bonnet will 
visually embody the abstract concept 
Liberty, however little connection there 
may be apparent between idea and symbol. 
The Greek used more thought and was more 
conscious of his aim. He used non-repre- 
sentational matter such as letters and 
legend very sparingly and succeeded in im- 
parting a surprising amount in purely 
visual terms. On the coins of the various 
ancient cities the geographical situation, 
the historical and mythical events of the 
past, often the very names of the towns and 


The in- 
must be re- 
thought in 




the value of the coin were all conveyed with- 
out a single written word. 

Syracuse showed its spring of pure 
water on a sea-girt island (the fountain 
Arethusa on Ortygia) as a young girl's 
head encircled by dolphins. The lake into 
which near Camarina a river widened be- 
fore it reached the sea was shown by a 
nymph upon a floating swan surrounded 
by a scroll of little waves, while on the 
other side of the coin a horned river-god 
with a sea-fish below him told the rest of 
the geography. At Selinus the story of 
Empedocles' success in draining the pesti- 
lential swamp was converted into terms of 
gods and bulls and marsh-birds and other 
simple visual apparatus. At Syracuse a 
four-drachma piece had a four-horse 
chariot, and two horses and a single horse 
indicated the two-drachma and the one- 
drachma pieces, while a wheel, as part of a 
chariot, was used for subdivisions of the 
drachma. There were no inscriptions 
stating the current value of these coins. 

The frequent "punning" or "canting 
badges," such as the rose for Rhodes, the 



goat for Aegae, were due to the desire to 
set even so unvisual a thing as the name of 
the town into representational form. 

In order to succeed in so thorough-going 
a conversion into visual terms, art obvi- 
ously must have recourse to symbolism; 
and to this extent Greek art became sym- 
bolic, since its representata often stood for 
things other than themselves as when a 
swan and nymph stand for a lake, a man- 
headed bull for a river, a marsh-bird for a 
marsh, or an object for its homonym (like 
Rhodes and the rose). But there was no 
mystic or mysterious intention and none of 
that deliberate obscurantism with which 
symbolism has so often allied itself. The 
symbolism of Greek coins was a kind of 
picture-writing; but it was not a hieratic 
script with an esoteric meaning for the 
initiate (as Christian iconography was at 
times), nor was there anything arbitrary 
and inevident in the relation between sym- 
bol and meaning, as tends to be the case 
when symbolism has literary or ritual as- 
sociations. It was not in an effort to point 
at mysteries or to obscure the obvious 





but for no 





with pretended subtleties, that Greek art 

had recourse to symbols. 

The Greek taste insisted on a thorough 

fusion of intellectual and artistic content, 

so that neither should obtrude itself to the 

detriment of the other. The Arethusa 

heads on Syracusan coins owe nothing of 

their artistic excellence to an understand- 

ing of their reference to a sea-girt island; 

nor is the simple picture-language ob- 

scured by irrelevancies added to improve 

the design or increase the artistic appeal. 

This canon is markedly inapplicable to the 

arbitrary and conventional symbolism of 

mediaeval iconography, where not to be 

initiate in the meaning is often to miss the 

artistic intention. The Greek artist was 

evidently no mystic: knowing clearly what 

he was about, he merely strove to make his 

intention equally clear to others. 





Having reduced the material content of 
his design to simple animate visual terms, 
the Greek coin-engraver further limited his 
range by making these terms conform to 
the material medium and to the shape and 
size of the field and background. Coin- 
stainping was a kind of relief sculpture in 
silver and gold, and no visual conception 
was acceptable unless it was appropriate to 
such an art. A town near a lake is in itself 
a visible thing, which can be shown by 
drawing; but it was unfitting to those 
formal artistic demands which coin-design 
suggested to a Greek artist. The nymph 
and the swan were animate equivalents, 
sculptural subject-matter, amenable to 
spatial arrangement within a circle as a bal- 
anced pattern with ordered lines and sur- 
faces. They had not merely an intellec- 
tualised meaning as symbolic of a god-held 
(and therefore man-beneighbored) lake, 
but an artistic import, in that they em- 
bodied artistic form and evoked artistic 
emotion. Heads, rather than full-length 

Our first 
Greek art 
was very 
well aware 
that art's 

province is 
the repre- 
sentation oi 
things. We 
now ask 
how this 
tion was 




igures, became favorite coin designs, be- 
cause the small size of the coin admitted 
sufficient detail for such a theme (but not 
for a larger one) while the metal was espe- 
cially fitted for those contrasts of smooth 
surfaces and chased areas which the sub- 
ject occasioned. And, finally, there was 
an inherent harmony of shape between 
coin and head, combined with an effective 
contrast between the unvaried curve of the 
one and the changing profile-lines of the 
other. With us the presence of heads on 
coins is rather an empty tradition: with 
the Greeks it was a discovery of the in- 
herent fitness of the theme to all the 
artistic requirements which could legiti- 
mately arise. It became canonised, be- 
cause of its artistic Tightness, in days when 
no mortal except the distant Persian king 
yet dared to put his own image on a coin 
to mark his own greatness and authority. 
It has survived for reasons of political and 
dynastic convenience rather than through 
any feeling for the singularly happy solu- 
tion which it offers to a difficult artistic 




We have considered various impulses 

and interests which determined the choice 

of artistic subject-matter in Greek art. 

And it has been indicated that the artistic 

process did not consist wholly in servile 

imitation of seen appearances to the best of 

the workman's technical ability. On the 

contrary, the representational subject- 

matter is presented under the modifications 

imposed by artistic forms. Art has its 


own devices of harmonisation of part to 

leads us 

part, its own devices of spatial suggestion, 

into a new 
field of 

its own devices for arousing and appealing 


to the esthetic sensibilities; and to the 

dictates and demands, of these devices the 

representational subject-matter must con- 

form. Our study passes therefore from the 

subject-matter of Greek Art to the Forms 

of its Artistic Presentation. 








AN often used (but perhaps not over-used) 

analogy speaks of art as though it were a 

language. In place of sounds art uses 

seen appearances and arranges these ac- 

cording to its own syntax and grammar in 

order to convey its artistic meaning. The 

analogy is inexact because the sounds which 

we use as words are nothing in themselves, 

whereas the sights of art are imitations of 

independently existent objects. Never- 

theless the comparison has the very great 

value of diverting the attention from the 

mere representational subject-matter of art 

and directing it toward non-representa- 

tional form. As this is the initial step in 

esthetics which the general public is so loth 

and so slow to take, there is no better 

philosophic approach than this metaphor 

or comparison. 

Under this analogy, the representata of 





art do service as the words of the language. 
The formal arrangement is the grammar 
and syntax. The esthetic emotion is the 
meaning. In order to impart this emo- 
tion, the artist puts representata in artistic 
form very much as we put sounds which 
are words in coherent grammatical con- 
struction in order to impart intelligible 
information. Under the same analogy, 
just as we cannot make a sentence with- 
out syntax, we cannot make art without 
artistic form; and just as a sentence with 
correct grammar may yet have only an 
idle or empty meaning, so representata in 
artistic form may have only a trivial or 
empty emotional value. Accurate repre- 
sentation through painting or drawing, and 
accurate imitation through modelling, will 
not in themselves constitute art, any more 
than a random picking-up of words con- 
stitutes a use of language. 

The representational subject-matter, 
moulded to artistic forms, fuses its imita- 
tive content with their non-representational 
appeal; from this fttsion results the mate- 
rial of our esthetic contemplation. The 

A primary 
of great im 
for the 




mind, trying to understand what it is that 
the eyes are conveying to it, is at the same 
time affected by the purely formal emotions 
of linear structure, pattern, balance, and a 
host of half-unclear suggestions and ap- 
peals; and all this emotional stimulation 
fuses with the recognition of the object 
represented. On the appropriateness or 
unappropriateness of the formal emotion 
depends the whole colour and feeling of our 
recognition. Indeed, the fact that our 
recognition is more than a mere identifying 
and naming of the object, that it is an 
esthetic perception with a distinct emo- 
tional colouring, seems to be due prin- 
cipally to two causes: (1) the knowledge 
that the represented object is an illusion, 
and not a real object which must be 
treated as we treat real objects in our real 
world, and (2) the fusion of representa- 
tional illusion with something non-repre- 
sentational and yet emotionally appro- 
priate to the representation in which it 

Our argument is much in need of con- 
crete illustration; and for this purpose a 



very well-known statue from the opening 
years of the fifth century will serve ad- 

With so generous an admixture of Thor- 
waldsen's restorations in them, the pedi- 
mental figures from Aegina cannot be used 
without caution. But the kneeling archer 
with the lion's-head cap, the so-called 
Herakles, has suffered less than most of 
his comrades and is apparently original 
save for his right forearm and the lower 
part of his left leg. The pose and the out- 
line are therefore assured, and Furt- 
waengler's investigations leave little doubt 
as to the angle from which he was intended 
to be seen. It can scarcely be accident 
that from this point of view, as from no 
other, the figure patternises to an intensely 
formal presentation in which an almost 
diagrammatic indication of the mechanics 
of an archer drawing his bow dominates 
the pattern. In the shooting of an arrow 
two forces are prominent, the strain of the 
bow-string back-drawn and the impetus of 
the released arrow on its flight. The 
strain and the direction-of -flight are nearly 

An example 




at right angles to the body of the archer. 
To offset this strain, a bowman will com- 
municate its thrust through some diag- 
onal prop or pose of his legs. Were we to 
make a diagram of the mechanics of an 
archer drawing a bow, it would look almost 
like a diagrammatic simplification of the 
sculptured figure of Herakles. " Natu- 
rally" (one might say), "because if those 
are the actual mechanics of a man shooting 
a bow, the archer cannot help displaying 
them." But the truth of the matter is 
that he can very easily avoid displaying 
them, for the mechanical strain and 
counter-adjustment are within, and not 
along the surface line of his limbs and 
clothes. That the pose can be shown 
without right angles or supporting diagon- 
als is patent from numerous drawings of 
bowmen which can be found in modern 
illustrations. But in the Aeginetan war- 
rior every line that is not suggestive of 
these forces is rigorously altered or" sup- 
pressed. Even the ornamentation of the 
jerkin is made of squares and right angles 
above and diagonal folds below, as though 



to force the eye into picking up that no- 
tion. Thanks to the lion's-head cap, even 
the head-form of Herakles is square. And 
what could better evoke the emotion of the 
straight, swift flight of an arrow when the 
string is loosed, than the long, forward 
stretch of arm with the flickering motion- 
line of its contours? 

Of course the artistic intent is not to 
present a dynamic graph of the equilibrium 
of forces: art is not interested in the in- 
tellectual presentation of physico-mathe- 
matical information. The purpose of the 
device is to awaken in the spectator a sense 
that there are such forces at work, in order 
that what he sees may not be only a casual 
picture on his retina but may come to him 
as an apprehension of a thing muscularly 
alive and full of powerful forces. While 
we are recognising the carved and coloured 
marble as Herakles the Bowman, we are 
at the same time being affected by purely 
formal emotions; and this emotion fuses 
with our recognition to produce the whole 
emotional state of our artistic contempla- 



The prim- 
ary distinc- 
tion again 

Pure forms 
per se are 


In the statue of Herakles the right angles 
in which the lines and masses meet do not 
in themselves represent anything, any 
more than a geometric theorem repre- 
sents actual objects (however much actual 
objects may exemplify and embody geo- 
metric theorems). The right angles are 
a pure form: pure, because we are consid- 
ering them apart from any material con- 
tent or representational significance; form, 
because they are an abstract schema into 
which representational matter may be 
fitted, as the kneeling man is fitted into the 
abstract pattern of lines. 

What is the emotional value of this ab- 
stract pattern? That would be hard to 
say. It is possible to make a 'drawing of 
such an underlying pattern without sug- 
gesting the silhouette of a human figure, 
and any one may experiment on himself to 
see whether he derives any emotion from 
contemplating such a pattern (provided he 
is able to look at it without thinking of it 
as a picture of a kneeling man or any other 
recognisable object of the real world). My 
own experience is that I have no clear or 



definite emotions from such a test. If I 
were to assert that I derived feelings of 
solidity, maintenance of equilibrium amid 
disturbing forces, power, endurance, or any 
similar sensations, I should be quite dis- 
honest with myself and be doing consid- 
erable violence to my real state of mind in 
forcing myself to believe that I was having 
any sort of definite, vivid, tangible emo- 
tion. But that I am affected (however un- 
analysably, however vaguely, as though by 
a dream without characters or scene) I 
could honestly maintain; and as the lines 
and surfaces are changed and the pattern 
changes with them, I can appreciate that 
my affection alters pari passu. But it is 
only when this abstract play of lines and 
angles and surface-shapes appears incar- 
nate in recognisable objects derived from 
the real world of my experience, that it 
seems to get sufficient emotional focus and 
bearing for me to appreciate clearly (or 
even describe) its character and range. 

This is the reef on which non-representa- 
tional sculpture and painting strikes and 
founders. In very recent times there have 


and there- 
fore cannot 
stand alone 




been attempts at a purely "dynamic" art 
in which we are asked to apprehend merely 
the emotion of surfaces, the clash of forces, 
the strife of line, the delights of linear 
motion, the appeal of contrasted and 
mingled colours. But no matter how we 
educate our sensibilities, we can never ex- 
perience anything tangible or definite out 
of these disembodied pure forms. The 
play cannot go on without characters. 
The spiritual kernel of a drama may be 
very abstract and lofty and vague and 
universal; but unless it embody itself in 
concrete human actors who speak intel- 
ligible words, those spiritual powers will 
never recreate themselves in the minds of 
an audience. Representational art is rep- 
resentational not by accident but because 
only so can it have its full effect. How- 
ever, this effectiveness does not lie in the 
mere picturing of objects from the real 
world, but in a successful fusion of this ob- 
jective matter with the formal suggestions 
which are the province of the particular 

Seldom is the reason for the appropriate- 



ness of a particular pattern to a given 
work of art so manifest as in the Herakles 
from Aegina. And yet most of us, if we 
are sensitive to the appeal of sculpture or 
painting, are aware and can feel (even 
though we cannot assign a reason for our 
impressions) whenever a pattern or play of 
line contributes to the emotion with which 
we contemplate an instance of those arts. 
In fact, it is a plausible theory that one 
essential distinction between esthetic and 
ordinary contemplation is the appreciation 
of abstract formal values in the field of 
vision and the fusion of these with the 
normal process of recognition of the ob- 
jects, so that there results an emotional 
(instead of a merely pragmatic or prac- 
tical) apprehension. If this is so formal 
values being accidental in Nature and not 
generally looked for by the spectator, but 
deliberately chosen for their influence and 
appropriateness by the artist and intro- 
duced into his work it should be obvious 
why esthetic contemplation may be emo- 
tionally so much more intense when it is 
directed toward a "copy of Nature" such 




as a work of painting or sculpture, than it 
is when in the presence of Nature itself, 
from which all such art confessedly derives 
its material themes. 




I have referred to the modernist sugges- 
tion that pure forms might be created 
without any representational content. It 
is worth a moment's attention, since it will 
lave an ultimate bearing on our theory of 
Greek art. 

In a certain London studio I was 
once shown an inlaid table-top whose 
geometric assortment and arrangement of 
planes and lines were intended to give me 
(so I was told) emotions of speed and power, 
of thwarted effort, and energy ready to 
burst forth. But I stood dully by and felt 
none of this intarsiate vitality rush over 
me. For it is not abstract speed and 
power that I can understand, but the speed 
of a railway train or the power of a goaded 
ox. Nor does the flying apart of six-inch 
lines give me a sense of flight except on a 
six-inch scale; but if I have the illusion 
that those lines are somehow (let us say) 
morsels of clod and bridge-rail, something 
of the sundering energy of a bursted shell 
comes vividly before my senses. The 





but must 
he fused 
with repre- 

The same 
in another 
Pure forms 
in painting 

table-top was a demonstration in pure form. 
It was also a demonstration of the futility 
of such formal effects when they are not 
immanent in the illusion of sensuous ob- 
jects, amid whose time and space we put 
ourselves with that strange sympathetic 
power which we employ whenever we see 
pictures in a mere square of painted cloth. 

By considering the art of painting, it is 
still easier to establish our argument con- 
cerning the artistic function of pure forms. 

The most obvious and easily distinguish- 
able pure forms which painting employs 
are line, mass, colour relations, from which 
are occasioned motion, pattern, and 
rhythm, with their accessories balance and 
thrust. I call these pure forms because 
they are independent of the subject-matter 
of the picture. They could exist whether 
the picture represented anything or not, 
and it is irrelevant to their existence 
whether the represented objects are rightly 
or wrongly drawn. They are pure forms, 
therefore, because they are not representata 
but schemata into which representata may 
be fitted. They are abstract in the same 



sense that geometric figures or kaleido- 
scopic designs are abstract, having for con- 
tent no objects of the world of experience, 
governed not by fidelity to Nature, but 
formulable according to some intellectual 
requirement, expressible perhaps in mathe- 
matical terms. 

How can line in an actual concrete paint- 
ing be an abstract form? does it not always 
bound or enliven the objects represented? 
then where is the abstraction, where the 
irrelevance to the represented object? 

True; but suppose that we isolate vari- 
ous lines by removing them from a picture 
and putting them by themselves. For 
lack of context, we have no notion of the 
objects which these lines helped to depict. 
We are treating them as unrepresenta- 
tional lines, as lines that show nothing. 
Are they all emotionally alike? do we get 
the same feeling from every one? Not 
quite. We may be somewhat at a loss on 
being asked to derive emotions from them 
at all. The performance is artificial and 
savours of the supersensitive behaviour of 
the avowed esthete. But I do not suggest 



are not 
per se 



glutting a healthy appetite on a morning 
rose, nor pretend that the contemplation of 
a page covered with lines of varying curva- 
ture should yield any profound emotional 
state. My theorem is quite simple and 
rather banale. A straight line looks stiff; 
certain curves appeal to us as graceful; 
wavy lines have a restless effect; and more 
complex lines compound more complicated 
phenomena of the same general sort. That 
is all that I mean by emotions derived from 
contemplating lines very vague, rather 
undefinable impressions. Whether they 
are suggestions borrowed from familiar ob- 
jects which show that particular line, or 
reactions due to an instinctive appeal to 
our gravitational and muscular sense, or 
products of our intellectual appreciation 
of the relations of the various component 
portions of these lines, is a matter of 
psychological inquiry which is irrelevant 
here and of no particular importance for 
the contention. If to this vague and 
feeble degree we are prepared to say that 
lines are not emotionally alike, no matter 
whether we can define or clearly dis- 



tinguish our feelings about them, if we are 
ready to say that not merely do the lines 
differ one from another, but that we differ 
when we look at one line and another, 
however slight and however trivial we judge 
this difference to be then I have estab- 
lished the purely formal character of lines. 
But lines may also be representational. 
Such and such a combination of lines de- 
picts such and such an object a tree, a 
fish, a frog, the Virgin Mary. Does a line 
lose its formal value, the moment we see 
that it represents a real object to us? 
Just here is the fundament and base of the 
contention. From the fusion of the two 
aspects of a line its purely formal value 
with its representational quality arises a 
new thing which I call the esthetic or 
artistic emotion. This new thing, which 
may be surprisingly intense and vivid, is 
not discoverable either in the represented ob- 
ject per se or in the mere formal -value of the 
lines used. It is a product of the fusion, 
often as unexpected and as novel as a 
chemical reaction. I admit the miracle, 
but I plead the fact. It is no more remark- 





and there- 
fore cannot 
stand alone 

able than the extreme intensification of 
emotion attendant upon the fusion of 
language with the recurrent beat of met- 
rical rhythm on which poetry draws so 

The forms of art, considered in and for 
themselves, are nearly always trivial and 
irrelevant. The esthetician, knowing this, 
never judges them thus for what they are, 
but for what they can do. Surely we might 
hold that the jingling of words of similar 
ending and the constant ordering of words 
into alternatingly accented and unaccented 
syllables is a puerile pastime for a grown 
man; but if we proceed to rule them out of 
poetry on that ground, we shall soon find 
out where we stand or some of the modern 
formless poets will soon instruct us. 

There is another formal function which 
line may perform that of suggesting mo- 
tion. The eye has a tendency to follow 
lines and to travel with greater or less ease 
according to the kind of line. It would be 
childish to claim that any very extensive 
spiritual experience is derived from thus 
travelling around on the lines of a picture. 



If we concentrate on the purely formal ele- 
ment, the result is extremely trivial. But 
if all these effects of acceleration and re- 
tardation, continuity and disconnection, 
intricate volution and open sweeping prog- 
ress be encountered and scarce-consciously 
performed during our contemplation of the 
objects presented by the picture, the triv- 
iality vanishes. The things in the picture 
take to themselves the emotional qualities 
which are latent in the pure forms. Swift- 
ness to our eye in travelling over lines is 
indeed great swiftness when those lines, 
measurable in inches, appear to us as wide 
uplands stretching to remote hills. And 
when difficult progress over broken lines, 
around uncomfortable angles, and through 
perplexing interlacings, is performed along 
what we accept for beams and joists and 
stays of a vast dungeon-room, our whole 
being seems confined and shut in, because 
we have transferred our ocular perplexity 
and fatigue to our imagined corporeal exist- 
ence amid the presented scene. 1 

But (it will be said) if each of the lines 
which depict a given object has its par- 


but must 
be fused 
with some 




ticular formal value, the resulting emotion 
depends on the particular shape of the ob- 
ject and not on the artist. That is true, to 
a certain extent. It is difficult to alter the 
emotional value of the lines of a very stout 
woman or an Ionic capital. But the ob- 
jection will quickly be seen to be super- 
ficial. There are numberless angles from 
which an object can be drawn, and hence 
numberless modifications of the necessary 
lines ; interior lines can be modified almost 
at pleasure; the relations of the object to 
the lines and surfaces of the rest of the 
picture are at the artist's discretion; and, 
finally, the object can be "misdrawn" so 
as to give it lines which would not actually 
occur in the photographically correct de- 
lineation. To what extent is this artistic 
' ' misdrawing ' ' allowable ? That is scarcely 
the esthetician's affair; but I am inclined 
to think that the only limit which can be 
demanded is the limit beyond which the 
spectator ceases to recognise what the 
represented object is intended to be. But 
artistic taste will nearly always assign a 




limit much closer to the photographically 

Extreme painters the Outragists, if I 
may so dub them often depart very 
widely from Nature. I must confess that 
to me distortions and malformations of 
decent human anatomy invariably intro- 
duce a strong element of displeasure and a 
revulsion away from all sympathetic con- 
templation, so that my final emotion is 
strongly modified by these unfavorable ele- 
ments. Now it is a matter of experience 
that wherever dislike and repulsion are 
markedly present as components, the re- 
sultant esthetic emotion is not likely to be 
of much value. (There must be a fascina- 
tion in ugliness before it can successfully 
heighten our artistic emotion . ) Our friends 
the Outragists would have us concentrate 
purely on formal considerations and forget 
the unpleasant modifications insofar as 
they touch the real world. We are not to 
think how we should scream if we en- 
countered in the open a woman with cubical 
hips and a mouth curling vaguely beneath 
one ear. " Forget all such reflections," 

Must the 
pure forms 
and the 
tions be 




tages of in- 

they might bid us; "this is not Nature, but 
Art. These lines and surfaces are ex- 
pressive, significant. They introduce the 
potency of abstract values conceived by 
the artist and undiscoverable by the an- 
atomist or the photographer. Something 
new has been created, a new and more 
primal emotion. To that lay yourself 

A new emotion, truly; but of what sort? 
Into the fusion of pure form with repre- 
sented object there enters a host of com- 
ponent elements. The represented object 
which here goes into the crucible is not 
just woman in general, but this particular 
woman with the bodily characteristics with 
which she is depicted, and along with these 
go all our associations, our attractions and 
repulsions, our memories and imaginations 
that are awakened by the sight of this 
woman as she is presented to our sight. 
There is no way of keeping these out of the 
crucible; for art is not an intellectual or 
geometric abstraction from ordinary ex- 
perience, but a transformation of our own 
sensuous world with all its inimitably com- 



plex ramifications of feelings, pleasures, de- 
sires, fears, superstitions, instincts, mem- 
ories, associations. All these go into the 
crucible. And if the brew be full of un- 
pleasant and distasteful humours, so that 
the final draught is bitter and puckers the 
lips, it is idle for the artist to bid us con- 
centrate our palate on some one flavour 
and ignore all the rest. 

As long as the resulting emotion depends 
on a fusion of form and matter (and it is 
our contention that this is the case) we 
must be prepared to recognise that the 
"new emotion" is an emotion about an old 
world with which we are irrevocably allied 
for our delights and our dislikes; and as 
long as we suggest that world by repre- 
senting its objects, we must be prepared 
for all the worldly suggestions that such 
representation entails. We can "mis- 
draw" as much as we like only, we must 
be prepared to accept the consequences. 

If we make this reply to the initiate mod- 
ernist, he will tell us that we ignore the 
special intention by being insensitive to 
the formal values. If we ask him then, 





"Suppose that, in order to introduce cer- 
tain formal values, the object came to be 
represented like (forgive the levity) an 
elephantiac scrub-woman recovering from 
a railway accident, do you hold that these 
formal values are arising around Woman- 
As-Such or around that particular dis- 
tressing specimen of her sex?" From this 
point the discussion should become more 
entertaining. For our contention must be 
that either the fact that a woman is repre- 
sented is irrelevant (in which case, why a 
woman at all? why not pure pattern?) or 
else this particular woman, just as she 
appears in all her discouraging deformity, 
is an integral part of the emotion. For 
that a painter can introduce "woman 
in general " into his painting and draw her 
or misdraw her as he likes without affecting 
her "womanliness" or our conception 
thereof, I entirely refuse to allow. The 
extremist idea seems to be that so long as 
we recognise what the object is intended 
for a house, a tree, a human being the 
formal values will fuse with our general 
concept of house, tree, and human being, as 




though the whole process went on iri a 
region of abstract thought instead of amid 
our immediate vision of particular objects 
every detail of which was effective. 

* * * 

Pure form to the detriment of repre- 
sentational fidelity, or representational 
fidelity to the detriment of pure form 
both are esthetically mistaken; for both 
tend to suppress an essential factor of the 
artistic appeal. 





A new step 
in the argu- 
Spatial pre- 
sentation a 
index or 
criterion of 
pure forms. 

We have insisted that there must be a 
sufficient measure of representational il 
lusion; but it is also equally clear that 
thorough-going representational accuracy, 
because it tends to eliminate the artistic 
form, cannot be a criterion of artistic 
achievement. In representational art 
actual sensuous deception is rarely in- 
tended. By imitation, the , appearances of 
the real world are shown in artistic form; 
but it is not art's desire to have that 
imitative presentation actually mistaken 
for objective reality. Of this we can read- 
ily convince ourselves by examining the 
various devices of spatial presentation 
which the arts employ. Often we shall 
find that the presentation is one-dimen- 
sional in its emphasis (insofar as there is 
an insistence on the linear aspect, with a 
consequent suppression of the more solid 
and extensive spatial qualities). Of this 
class are many of the Greek vase-designs, 
especially during the Early Red-Figure 
period; and one essential element of Greek 




low-relief is this insistence upon linear 
presentation. In other artistic products 
the presentation is two-dimensional in its 
emphasis (in that there is an insistence 
upon the surfaces which the objects of 
esthetic contemplation occupy in the field 
of vision, with a consequent diminution of 
the illusion of solid spatial existence). Of 
this class are certain vase-designs, much 
relief-carving, and (presumably) most of 
the earlier wall-paintings. Here too we 
shall have to classify the essential element 
in the esthetic appeal of Greek architec- 
ture. Finally, there is three-dimensional 
presentation, the cardinal problem of 
sculpture-in-the-round and one of the 
great preoccupations of architecture 
(though not in classical Greece) and of 
painting (in certain very developed periods 
of its career). Since successful three- 
dimensional presentation occasions a very 
vivid impression of completely objective 
existence in space, and since three-dimen- 
sional presentation alone can achieve this, 
it is only in the arts which aim at such 
presentation that imitative illusion ever 

and three- 



There are 
other pure 

'orms and 

methods of 

jut we will 
our argu- 


threatens to be complete and to become 
delusion. Against such a fatality sculpture 
usually guards itself by an insistence on its 
material medium and the constant dis- 
crepancy between what actually is and 
what by imitation appears to be. Painting 
averts a similar mischance by creating its 
own luminous environment in which its 
imitations are kept within a spatial con- 
struction of its own, which will not fuse 
and therefore cannot be confused with the 
objective space in which the beholder 
stands. Architecture, content to imitate 
conventional objects of its own choosing, 
has no need to seek to avoid the appear- 
ance of solid reality for its creations, but 
on the contrary strives to present them in 
their spatial actuality with all vividness 
and a heightening insistence, achieving for 
itself almost the distinction of a fourth and 
unique genus of spatial presentation, the 
presentation of solids enclosing space. 

It is to these devices of spatial presenta- 
tion to which particularly this study will 
turn, rather than to the more familiar and 
well-understood formal devices such as 




composition, balance, "mass," " tonality," 
rhythm, and the other commonplaces of 
ordinary artistic appreciation. 

* * * 






Greek art 
uses one- 

In early fifth century Greek vase-designs 
the representational subject-matter is pre- 
sented seemingly without desire to impart 
an illusion of spatial depth or solid exist- 
ence in space.. The draughtsmen of the 
early red-figure period turn their attention 
to the discovery of linear equivalents for 
the objects which they purpose to draw. 
The gods and heroes, athletes and revellers, 
whom they present to our sight, exist 
mainly by virtue of the encompassing lines 
which mark them off from their back- 
ground. Internal lines are used for de- 
tails of muscular grooving and drapery 
folds; but these lines are not used to 
create plastic illusion, they do not model 
or occasion any apparent spatial pro- 
trusion or recession, toward or away from 
the eye. The world of real objects is re- 
duced to terms of linear appearances, and 
other spatial qualities have evaporated or 
disappeared. With their disappearance a 
subtle and barely describable change at- 
taches to our impressions of the world in 




which these figures move the world of 
these legendary people and athletes and 
revellers to whom we somehow allow exist- 
ence as though they were real and alive, 
and yet whom we never mistake for flesh 
and blood. In our impressions of them we 
scarcely admit that they have the weight 
which belongs to solid material objects, nor 
that they must exert effort in order to 
move themselves. We ascribe to them a 
life of extraordinarily heightened anima- 
tion and know somehow that the impulse 
of movement rules and runs through them, 
holding ankle and knee and hip and shoul- 
der and arm together in one control. 
These are perhaps but vague and partial 
descriptions of what we feel; yet they will 
serve to suggest that the presentation of a 
human body wholly through linear equiv- 
alents what I should like to call a presen- 
tation with one-dimensional emphasis 
carries with it a power of peculiar sugges- 
tion and that this power is one of the spe- 
cific esthetic resources of any art which 
chooses to employ such means of presenta- 




As long as we are foolish enough to be- 
lieve that realistic illusion is the end and 
aim of art, we shall be wholly unable to do 
justice to the intentions of such an art's 
more purely linear methods. But if we are 
prepared to see in esthetic phenomena a 
fusion of objective representation with 
artistic forms capable of transmuting (even 
to the point almost of destroying) the 
sense of objective reality in the interest of 
artistic emotion, we shall have little dif- 
ficulty in understanding the devices of 
Greek vase-design. 

For example, we shall readily appre- 
ciate that it was not so much because the 
pigment dried rapidly (as is frequently 
asserted) that the ancient vase-drawings 
show outlines which must have been drawn 
quickly, continuously, and without hesi- 
tation, but, rather, because the whole ob- 
jective existence of the portrayed figure and 
all those impressions of. which we have just 
spoken depended on these lines. An in- 
terrupted contour, an incoherency of linear 
connection, would lessen the power of these 
impressions and work against the very pur- 



poses for which linear presentations were 
employed. The eye must be made to fol- 
low the run of contours and inner lines, so 
that the mind may never rest in the illu- 
sion that the pictured figures actually fill 
and occupy the space in which they are 
shown. And further, all suggestions of 
motion and those impressions consequent 
on the travel of the eye along lines, depend 
for their vividness on the clarity and insis- 
tence with which the linear exceeds any 
other spatial emphasis. 

On the white-ground Attic lekythi later 
in the fifth century, although the method of 
presentation is still very linear, there is in- 
troduced the practice of making solid areas 
of color and of using these to build up area- 
relations in the field of the design. There is 
still no illusion of spatial depth; but the 
objective world is now no longer conjured 
up solely through linear equivalents. The 
eye, instead of moving continually under the 
incitement and to the lead of lines, now 
rests on the full areas of color. In fact, the 
pictured objects of this world are pre- 
sented largely in two-dimensional guise and 






with two-dimensional emphasis: they exist 
surely as surfaces, though still very pre- 
cariously and uncertainly as solids in space. 

Two-dimensional presentation very gen- 
erally avails itself of the device called pat- 
tern. This device is so familiar that it 
scarcely needs elucidation. The textile 
arts, the arts of decorative metal-work, in 
fact all arts whose representations are con- 
ventional and ornamental rather than 
directly illusionary, count much on pat- 
tern. Even relief -sculpture and painting 
admit it as one of their major pure forms. 

Now, what pattern is and why it appeals 
to us, I hope that I shall not be called upon 
to discover. For one thing, it depends upon 
symmetry of parts, a correspondence of 
right with left or upper with lower. A 
kaleidoscope cannot fail to produce pat- 
terns; for however orderless and crazily 
the central coloured sherds lie, the mirrors 
will so echo these elements by reflection and 
inversion that order and balance will be 
built up on disorder and there will be pat 
tern. And in apprehending pattern there 
is both the simple joy of recognition (when 



we detect repetitions and inversions of the 
same theme) and the intellectual joy of 
discovering spatial relationships and the 
generative principles of construction. But 
there must be some deeper appeal than this , 
for the sense for pattern seems to be more 
primitive and to lie deeper than such 
sophistications and the joy in it seems to be 
strangely clean and satisfying. Intellec 
tually we may feel contemptuous of the 
triviality and practical irrelevance of its 
behaviour. Yet something in the contempla- 
tion of pattern goes as deep as our instinct 
against bodily deformity and approval of 
fully and evenly developed things, whether 
they be flowers or children or our lovers' 
bodies, (is it then to be wondered at if pat- 
tern is insistently present among the artis- 
tic forms of a people so addicted to athletic 
training and the sight of the nude as were 
the Greeks?) 

Whatever this sense of pleasure and ap 
proval may be in itself, it is carried over to 
the represented objects of art when these 
objects are visually grasped as inherent 
components of a patterned design. Fur- 





thermore pattern unifies the objects to 
which it is applied and gives a consequent 
sense of internal cohesion, so that scattered 
elements of a landscape or discrete human 
beings of a sculptural group may impart, 
merely because of the pattern in which they 
are put, a sense of relevancy and fitness to 
appear together. We are aware of a unity 
which we may take for the all-penetrating 
unity of Nature or the 'unifying force of an 
artist's creative idea, though it arise merely 
from the unity of the unmeaning pattern 
in which the trees and fields and hilltops 
of a painting are shown. 

What art may do with a pure form, with 
pattern void of representation, may be well 
studied in Mohammedan art. The religious 
ban on imitating and reproducing the 
appearances of living things turned the 
craftsmen of Islam toward a half -mystical, 
highly geometric, and thoroughly interest- 
ing pursuit of the decorative possibilities 
of pattern. Rugs and carpets, whose orna- 
mentation is an Oriental tradition (our 
western methods of presentation went into 
the picture-world of tapestries) show a 



imilarly highly developed art based on the 
)ure form, pattern. Birds and trees and 
flowers enter in, but scarcely for their own 
sake. They are usually patternised in 
appearance, with no pictorial space-rela- 
tionship to one another or other setting 
than their appropriate position within the 
pattern. Form has the upper hand, repre- 
sentation is incidental. Such work, what- 
ver other artistic value it may have, seems 
to miss the spiritual opportunities of more 
representational design, because it does 
nothing with every-day human experience. 
It can appeal to certain deep-seated likes 
and pleasures and instincts; but it does 
nothing with them, it does not humanise 
them, it does not bring them to bear upon 
that world of sensuous life in which our 
spiritual experience is rooted. 

It would be possible to classify the Greek 
use of artistic pattern under the captions 
Decorative, Unifying, and Suggestive. Dec- 
orative patterns would be those in which 
the pattern contributes merely the pleasures 
attendant upon its orderliness and loveli- 
ness of arrangement. Unifying patterns 





would be those in which the synthesising 
power of the pattern is transferred and 
applied to the represented objects. Sug- 
gestive patterns would be those in which the 
geometric construction of the pattern 
conveys some suggestion of equilibrium, 
conflicting forces, or other mechanical 
effect. This last use is most common dur- 
ing the latter half of the fifth century and 
is conspicuous in the metope compositions 
of the great Doric temples of that period. 
The Lapith-and-Centaur metopes of the 
Parthenon are generally composed about a 
central polygon of uniformly coloured 
background; the enclosing sides of these 
polygons are the bodies of the two contest- 
ants ; a very br.ief analysis shows that pat- 
tern is here functioning in all three of the 
aspects just mentioned, and that truth to 
Nature, spatial, and all other illusions are 
rigorously subordinated and distorted in the 
interest of these three aspects of pattern. 
In consequence these metopes have been 
generally criticised adversely. I know of 
no critic or writer who has appreciated 
them for what they really are and for what 



they manifestly try to be two-figure 
compositions presenting intense physical 
strain of bodily contest wholly through the 
artistic forms of line, surface, and pattern. 
They were to be seen at a considerable dis- 
tance and at a height of at least forty- 
five feet above the ground in a position, 
consequently, where simplifications of line 
and pattern would be effective and minor 
fidelities to representational truth would 
be inconsequent 

Familiar as pattern is, artists and esthe- 
ticians seem seldom aware that it is wholly 
a two-dimensional form arid that therefore 
its occurrence in three-dimensional presen- 
tation is a matter of very unusual interest 
and importance. In a painted scene there 
is intentionally the illusion of depth: we 
construct a space in which the objects 
occupy positions at various imagined dis- 
tances from the eye. Insofar as we accept 
this space as actual, no pattern is possible; 
we only apprehend the pattern by reducing 
all the objects to that which they really are 
flat surfaces of colour lying" in a single 
plane. One peculiar effect of pattern in 





such a painted scene, therefore, is the de- 
mand which it makes on us to apply a two- 
dimensional arrangement of coloured areas 
to a three-dimensional spatial construction 
and (impossible as it may sound) recognise 
that this space without depth and the 
space with depth are somehow fused, some- 
how one and the same. It should follow 
that, the more insistence there is on pat- 
tern, the less actuality we ascribe to our 
construction of a three-dimensional space 
in which the objects stand; and this might 
be carried to such a degree that we should 
be so dominated by the sense of pattern as 
to be wholly unable to make a three-dimen- 
sional construction at all. 

I hazard the conjecture that this was 
actually the case with Greek wall-painting 
during the fifth century because the empha- 
sis was almost entirely on the two-dimen- 
sional devices of area! composition (pat- 
tern, balance, repetition of similar shapes) 
and on the one-dimensional devices of 
linear presentation. 2 But of the actual 
effect of Greek wall-paintings we know so 
little that it will be wiser to confine the dis- 



cussion to a kindred art whose products 
are abundantly preserved to us. 

Carved relief in Greece, notoriously, is 
subject to a seemingly arbitrary convention 
of execution in planes. Instead of a uni- 
form abbreviation of depth (according to 
which, although two or three inches of pro- 
jection may have to do duty for the entire 
depth of a represented scene, this projec- 
tion will be correctly distributed pro rata, 
and every element will get its proper share) 
in the Greek convention of planes there is a 
very uneven distribution of depth. A 
human torso will be carved as an almost flat 
surface with undercut outline; and in 
general all the depth will be concentrated 
at the contours, within whose boundary all 
will be treated practically as a continuous 
plane-surface. A succession of planes, each 
stepped back at its edges to the next plane, 
will make up the entire relief. Of this series 
of planes, two will be much more prominent 
than the rest : the front plane, which agrees 
with the original smoothed surface of the 
marble slab, and the rear plane, which is 





the (uniformly coloured) background against 
which the figures appear. 

The origin of this convention is well 
known. Relief originated in Greece in the 
practice of cutting away the background 
from a drawn or painted figure, leaving a. 
flat drawing raised above a flat background. 
Where elements in the drawing overlapped 
or were intended to be at different distances 
from the eye, the farther element was set 
back from the hither one by the same device 
of cutting it back a stage. Out of this prac- 
tice there arose quite naturally the conven- 
tion of composition in successive planes. 

To have understood the origin of this 
convention is a very different thing from 
comprehending its artistic validity once it 
was established. If the Greek sculptors 
adhered to this convention throughout the 
fifth century, it is only fair to assume that 
they did so because they found it artisti- 
cally useful and not because they were 
unable to hit upon any different method. 

The frieze of the Parthenon presents the 
Pan-athenaic procession so divested of all 
scenic setting and of all recognisable spatial 



surroundings that archaeologists have ser- 
iously asked the question whether it is the 
real procession or some sort of a rehearsal 
which is shown. Riders and chariots, 
elders and musicians afoot, youths with 
animals, maidens with sacrificial vessels, 
all move as though on a narrow shelf, seen 
sharp against the sky. The horsemen are 
often three figures deep, their arrangement 
even demands that they should be inter- 
preted as riding six and seven abreast; yet 
space has somehow lost its density, and we 
fail to see them in the space which we know 
they must occupy. There is an abstraction 
of all else, and an insistence on our sense 
for harmonious and effortless motion, 
which gives to the contemplation of the 
Pan-athenaic frieze a quality of delight such 
as no actual procession may give. The es- 
thetic explanation is apparent. Thanks to 
the convention of planes, the presentation 
is largely two-dimensional. Depth is a 
half -unreal assumption, and of it we have 
only as much as will satisfy the require- 
ment of making motion in space a possible 
inference. Thanks to the convention of 





planes, we are forced to rely upon the con- 
tours to give these figures spatial existence; 
and so we are enthralled in the illusions of 
linear presentation. But also because of 
the convention of planes, the pure forms of 
two-dimensional presentation are given 
free rein. In the cavalcade the set recur- 
rence of horses' heads and riders' heads at 
the top of the frieze, the smooth areas of 
the horses' bodies at mid-height of the frieze, 
the rapid beat of recurrent horses' legs at 
the bottom of the frieze, can aftect us with 
their spaced rhythms to an extent which 
would be impossible with a stricter illu- 
sion of depth. For, as we have seen, all 
these devices of pattern exist only insofar 
as we are able to reduce the presented 
objects to areas on a single surface. 

[Until the last quarter of the fifth cen- 
tury (the frieze from Bassae is one of the 
earliest instances to the contrary) the men 
and animals on a Greek relief, if they seem 
to move, always suggest motion in the plane 
of the background and not toward or away 
from the eye. This was largely a result of 
the frontal presentation which early <1r;i\v- 




ing naturally employs ; but it had its artis- 

tic justification because the plane of the 

background is the only direction which does 

not demand the construction of additional 

spatial depth (since if there is to be motion 

we must necessarily imagine a space in 

which it can take place). But additional 

spatial depth will further the three-dimen- 

sional illusion and thereby obscure the 

effects incident to surface and line. 

The convention of planes and the con- 

vention of motion in the plane of the back- 

ground are therefore artistic devices which 

are involved in the problem of two-dimen- 

sional presentation; their presence and 

meaning in the developed art of the fifth 

century cannot be adequately explained in 

any other way. 

With sculpture in the round we at last 

and three- 

reach the problem of three-dimensional 


presentation, with all its attendant intri- 


cacies and difficulties. 

Not only is the question of three-dimen- 

sional presentation itself a difficult one;" but 

the sculptural appeal is rendered extremely 





complex and confusing for the esthetician 
because, just as the arts which emphasize 
two-dimensional presentation generally in- 
clude also the devices of one-dimensional 
emphasis (as when carved relief employs 
linear effects) so an art whose characteris- 
tic method of presentation is three-dimen- 
sional will impress both one- and two- 
dimensional presentations into its service. 
Since sculpture uses both the pattern de- 
signs of areas and the linear suggestions of 
contours and interior lines, there results 
for the sculptural appeal a fusion of many 
diverse factors and a consequent complex 
ity such as does not characterise the arts 
which we have thus far considered. 

As an instance of this synthesis of many 
factors, the relief of Aphrodite rising from 
the sea (if such be the correct description 
of the scene on the so-called Ludovisi 
Throne) shows a masterly combination of 
the suggestive emotional powers of all 
three forms of spatial presentation. By 
pattern the attention is centered on Aphro- 
dite; by linear suggestion of parallel cate- 
nary curves, on which as it were the figures 




are hung, a sense of weight in effortless 
equilibrium is conveyed; by a heightening 
of the projection of the relief from its back- 
ground toward the centre and a gradual 
lessening of it toward the right and left 
and bottom (where there is practically no 
spatial construction, but only linear pre- 
sentation) Aphrodite herself is presented 
with an illusion of spatial actuality which 
does not obtain for her two attendants, so 
that she, herself bodily real, is surrounded 
by a sort of penumbra of unsubstantiality. 
And this works magically upon our emo- 
tional comprehension. 

In this extraordinary effect of a fusion of 
corporeal solidity with utter unsubstan- 
tiality sculpture has a unique opportunity. 
Most easily accentuated in relief-work, it 
yet may exert a strong influence in sculp- 
ture in the round, so that coincident with 
a three-dimensional presentation there 
may be the eerie negation of all the solidity 
with which the eye is so directly and so 
insistently affected. From linear presen- 
tation almost all the sense of motion in a 
statue is derived (though in a more Cyclo- 




which leads 
us to a 
and subtle 

pean way it can be suggested by surfaces, 
as in the Torso Belvedere). On relation of 
surfaces depends much of the appeal to the 
mechanical sense, the sense for equilibrium 
and control of bodily movement. In order 
to be truly effective, both this linear and 
this areal presentation must be embodied 
in a three-dimensional presentation, where- 
upon the influences of these formal devices 
will be ascribed (illusorily) to an . actual 
spatial existent to which they will impart 
their peculiar emotional suggestions. It is 
in this fusion of one-, two-, and three- 
dimensional presentations that the real 
subtlety of the sculptor's art resides. But 
because of the complexity of the result, 
the esthetic analysis of the sculptural 
appeal has remained to this day largely 
unguessed and unwritten. 

It is indeed scarcely surprising that, 
while the esthetics of the arts of one- and 
two-dimensional presentations are compar- 
atively well understood and correctly stud- 
ied, the arts of three-dimensional presenta- 
tion have kept their secrets to themselves, 
so that sculpture has usually been con- 




sidered with the utmost naivete and archi- 

tecture has seemed a simple, open and un- 
elusive art whereas, in reality, because it 
combines with its own specific device of 
presentation (solids enclosing space) the 
peculiar appeals of nearly all the other arts, 
it affords esthetic phenomena so complex 
in their combinations as to bewilder and 

wholly mislead the esthetician. 
Accordingly, it will perhaps be excused 
as a not entirely disproportionate emphasis 
if the remainder of this study be devoted to 
a consideration of these two, esthetically 
the most highly complex of the arts, and to 
a scrutiny in considerable detail of the 
attainments and devices of Greek sculpture 
and architecture. 

to whose 
solution our 
is chiefly 








Spatial pre- 

THE cardinal esthetic problem in the analy- 

still our 

sis of the sculptural appeal is the determi- 


nation of the way in which statuary works 

and guide 

on our sense for spatial construction. We 

incline to view wax-works merely as arti- 

ficial human beings that might be flesh and 

blood if they did not happen to be wax. 

The aim is delusion; and, beside the patent 

fact of immobility, the only material differ- 

ence between original and imitation is the 

difference of material. We do not so see 

sculpture when we view it as a work of art. 

Effect of 

In order that we shall not so see it (as a 


realistic illusion) the art is careful to omit 

one of the essential marks of the real world 

its colour. Early Greek stone carving 

did not do this: it coloured brilliantly, 

but conventionally. During the succeed- 

ing periods the lavish use of colour may 

have gone gradually out of fashion. Cer- 




tainly, the Hellenic bent for representa- 
tions of the nude tended to keep the poly- 
chromatic influence subsidiary, and though 
the nude parts were somewhat altered in 
tint by the process which the Greeks called 
ganosis, this seemingly did not produce an 
actual illusion of human flesh; while in 
bronze there obviously could be no inten- 
tion of illusionary colouring (in spite of a 
few casual anecdotes of instances to the 

Gothic sculpture was in general highly 
coloured. The Renaissance (perhaps 
through the error of attributing a complete 
absence of colour to the surviving ancient 
marbles from whose surface time may 
merely have removed all traces of the tints) 
accepted a colourless convention almost 
without opposition. From that time on, 
the tradition of unstained marble has 
maintained an almost unquestioned ascen- 
dancy. To-day it is to be justified not as 
an accidental historical heritage, but on 
purely practical esthetic grounds, since it 
destroys the outright illusion and thereby 
facilitates the peculiar power of sculpture 





to work on us through spatial forms of its 
own devising. 

In particular, the blank unorbed eye, by 
which the modern sculptor so readily makes 
us accept his carven men and women as 
purely ideal equivalents for flesh and blood, 
was a device seemingly unknown in Greek 
times, when it would have been at variance 
with the essential trend toward naturalism 
which, as we shall see, dominated the suc- 
cessive generations of sculptors. 

In the archaic period there was no ques- 
tion of using colour to produce a material 
illusion. Where the hair of the head and 
the beard could be painted blue or clear 
bright red, we need scarcely raise the ques- 
tion of the ulterior purpose of such decora- 
tion. In the early Hellenistic Age, when 
the Alexander Sarcophagus was made, the 
realistic illusion was much heightened; 
yet, even then, the tones seem to have been 
so clear and pure that we may doubt 
whether anything but a very modified 
llusion of real appearances was intended. 
In view of the realistic attainments of the 
period, it is not safe to insist upon this 



Doint. More certainly, colouring was still 
lalf conventional in the period of Pheidias; 

t we are told on good authority that 
Praxiteles paid considerable attention to 
this aspect of the art, so that we may 
assume a more naturalistic tendency to 
lave set in during the fourth century. 

It would seem that insofar as the true 
colours of the real world were applied to 
Greek statuary the effectiveness of its 
:ormal qualities was not heightened but 
diminished. This would be a correct de- 
duction were it not for the fact that a 
judicious use of colour can enhance all 
those formal effects which depend upon the 
comparison and correlation of surfaces and 
areas (such effects as those of pattern and 
balance of masses), as well as emphasise 
outlines by contrasting the colour of adja- 
cent areas. If the colour-scheme be kept 
conventional, these advantages may be 
turned to account without the disadvan- 
tages attendant upon illusionary colour; 
and such a practice must have obtained in 
Greece before the conventional colouring 
inherited from the archaic period had 




Effect of 

been entirely abandoned, that is to say, 
in all probability, during the whole of the 
fifth and the first part of the fourth 
centuries B.C. 

To return to an earlier part of the argu- 
ment, the cardinal esthetic problem is the 
determination of the way in which the 
sculptural appeal works with our sense for 
spatial construction. The sculptor's aim 
is to present to us a cubic solid in such a 
manner that we can intimately appreciate 
its existence as a three-dimensional object. 
He must enable us to apprehend the solid- 
ity of solids directly by mere ocular- con- 

Ordinarily we employ a sort of intellec- 
tual construction toward the seen world. 
From the various appearances of objects 
from varying points of view we infer their 
cubic shape and position in space. We read 
off the meaning of foreshortening and per- 
spective, and so get at an understanding of 
the solidity of objects not directly, but at 
second-hand. But in the apprehension 
of the sculptured form we get a more com- 
plete and a more direct visual comprehen- 



sion of the solid object as a solid. We do 
not see all around it (that clearly is a geo- 
metric impossibility) and yet we are inti- 
mately aware of its further boundaries 
(which are the surfaces on the invisible 

The real difficulty in spatial apprehen- 
sion lies in the fact that a three-dimensional 
world is presented to our eyesight in what 
amounts to a two-dimensional form and 
we have to construct the missing dimension 
out of such data. The dimension in depth 
away from the eye is obviously the difficult 
one, and it is part of the sculptor's business 
to construct this for us so that we can appre- 
hend it without mistake or trouble. How 
can he do this? By removing as much as 
possible from the difficult dimension and 
showing it in the two easy ones. By loading 
the two dimensions, he lightens the third. 
In practice, this amounts to the employ- 
ment of three very definite devices: 
(1) intelligible pose, (2) planes of compo- 
sition, (3) "modelling" lines. 



and its 




Relation of 
outline to 


(1) By intelligible pose I mean simply 
that the model must be presented in such 
a position that the beholder can clearly and 
fully understand what it is that he sees. 
This seemingly simple demand causes the 
greatest possible amount of trouble, just 
because there are as many aspects of a 
statue as there are points of view from 
which the beholder can look at it. In the 
case of a free-standing statue in the round, 
it is apparent that theoretically there are 
as many points of view -as there are points 
on the circumference of a circle, and that is 
an infinite number. Practically, the 
spectator will not demand complete intelli- 
gibility from more than a very few stand- 
points, and may usually deem himself for- 
tunate if he can find any which are wholly 
satisfactory. Even with these limitations, 
it must be remembered that it is one and 
the same cubic form which determines all 
these more or less planilinear appearances 
(just as a hundred different photographs 
can be taken of a single solid) and a change 




occasioned in order to improve one aspect 
will introduce a relative change into all the 
other aspects. Yet we began by demanding 
that from every viewpoint the pose should 
be intelligible, so that the whole construc- 
tion of the represented object as a solid in 
space may be read off from any one of this 
series of restricted and quasi two-dimen- 
sional appearances. 

More concretely, we may demand that 
no foreshortening should take place if it is 
not immediately intelligible as a fore- 
shortening, so that we may apprehend the 
actual depth of the element which appears 

In order to meet this demand the sculp- 
tor may try to make the silhouette always 
so characteristic of the bodily parts which 
it outlines that the mere contours will make 
them intelligible. 1 In this way he will be 
crowding as much representational sig- 
nificance as possible into the two visible 
dimensions and leaving as little as possible 
to the invisible dimension. The device 
depends for its success upon the sculptor's 
ability to draw an expressive contour and 




upon his sensibility to linear equivalents 
with which to conjure up an objective 

The tradition of the expressive contour 
grew up and matured very naturally among 
the Greek sculptors. Primitive art draws 
very largely in profile, since the profile 
view is the most readily intelligible and the 
most vividly retained by the visual mem- 
ory. In draughtsmanship, foreshortening 
is a difficult accomplishment because the 
outlines which it demands are not those 
which the mind insists upon retaining as 
characteristic. Sixth century vase -designs 
are all based on profile appearances. This 
was also true of early relief, because the 
artist proceeded from a drawing sketched 
upon a smooth surface of stone and chipped 
away the background 'around this design. 
More surprisingly, even sculpture in the 
round is subject to similar influences. The 
sculptor strove to embody the human form 
as he visualised it. In the case of marble 
statuary he may have drawn a fiat image on 
the face of the stone and then hewn inward 
for the third dimension until he had given 



his sketch a solid embodiment. The early 
statues are thus calculated from a single 
position, from immediately in front. Their 
depth was apparently controlled from 
points at right angles, that is squarely from 
the right and left flanks of the statue. In 
consequence of such a procedure, the statue 
was evidently fashioned for the four cardi- 
nal viewpoints (front, back, right, left) 
and from each of these it displayed the 
characteristic silhouette which our visual 
memory especially retains. 

The Man with the Calf (late sixth cen- 
tury Attic) is an admirable example of a 
frontally composed statue. Seen from in 
front, it exhibits a pattern of line and sur- 
face such as a drawing or flat relief might 
show. As we move around the stone, this 
pattern gradually folds up and vanishes 
and no satisfactory viewpoint occurs until 
we come squarely on the flank, when a 
secondary but not wholly intelligible com- 
position arises, with profiles of the heads 
of the calf and its carrier. 

So marked is this frontality and so strik- 
ing is its two-dimensional pattern, that 



when we stand squarely in front it is almost 
impossible to see the statue as a solid. It 
seems flat, in one plane. In the terms of our 
previous discussion, the third dimension 
has been so drained of its content and the 
other two dimensions have been so filled, 
that under the influence of the plane-mak- 
ing pattern there is nothing to give the illu- 
sion of depth. The process has in fact been 
carried too far. The contour has been made 
so intelligible and the frontal plane has 
been so emphasised that the dimension of 
depth has no necessary function or obvi- 
ous existence. 

Such frontal composition has the patent 
advantage of insuring an expressive con- 
tour but not for all viewpoints. How a 
harmonious succession of expressive con- 
tours can be combined into a single solid, 
so that from every point of view the statue 
may be immediately intelligible, thus comes 
to be a cardinal artistic problem connected 
with sculptural pose. The Greeks solved 
it gradually. A century after the Man with 
the Calf, Polykleitos has not yet found the 
full solution. His formula (as embodied in 




the Doryphoros and the Diadumenos) 
shows only a frontal solution. The -rigid 
median line now bends in a long, gentle 
curve and counter-curve from ankle to 
neck; but this curve is all in one plane. 
The weight-leg and free-leg with the accom- 
panying raised hip and lowered shoulder 
are all vertical and lateral displacements, 
visible from in front but barely appreciable 
in the side view. Pliny complains that 
Polykleitan statues are too "square." 2 
Praxiteles inherited and exaggerated this 
formula, accentuating the curve but still 
keeping it in the frontal plane. A fifth or 
early fourth century statue may be imme- 
diately recognised and dated by this adher- 
ence to frontal composition. A century 
after Polykleitos, Lysippos is at last in 
command of the mysterious formula for the 
pose which shall yield expressive contours 
for every viewpoint. His refinements of 
pose are no longer merely frontal but pass 
into every plane, being based on a progres- 
sive rotation of the horizontal axes of the 
human body. Pliny's famous phrase 3 
about Lysippos, "nova intactaque ratione 




quadratas veterum staturas permutando" 
makes patent reference to this innovation. 
If we draw an imaginary line through the 
two ears, another through both shoulders, 
another through both hips, another through 
both knees, another through both ankles, 
we shall have the five horizontal axes. If 
we make this experiment on a statue of 
the period of the Man with the Calf, we 
shall find these five axes all pointed in the 
same direction. There is no torsion to such 
a perfectly frontal composition. If we make 
the same experiment on some of the early 
Hellenistic statues, we shall find that the 
axes gradually turn like the needle of a 
compass so that the lowest one points at 
right angles or even in opposite direction to 
the topmost. This gradual torsion carried 
through the whole body is the typical 
Hellenistic (or should we say Lysippic and 
Rhodian?) formula for producing an "om- 
nifacial" composition. Thanks to its 
gradual nature, the elements in intelligible 
profile always explain the adjacent ele- 
ments which are moving into foreshorten- 
ing. As the axes are often distributed 



through a right angle or more, it is likely 
that there will always be some element in 
profile. 4 From the employment of this for- 
mula comes the extraordinary complete- 
ness of each aspect under which the statue 
can be viewed (since each aspect is ani- 
mated by the same formula as all the 
others), and the -remarkably harmonious 
manner in which each aspect arises out of 
the preceding and melts into the succeed- 
ing one, as we move around the statue. 

Excellent illustrations of an extreme use 
of the formula are the Dancing Satyr from 
Pompeii, the Seated Hermes from Hercu- 
laneum, the "Narcissus" or "Listening 
Dionysos" (all three in the Naples Muse- 
um), and the Borghese gladiator in the 

Naturally, not all Hellenistic sculpture 
is composed to this formula nor is the for- 
mula always carried consistently through 
all the axes; the axis of thejiead is least 
likely to conform since there is no obvious 
artistic reason why it should. It is. no 
academic rule of perfection; yet it recurs 
with surprising frequency during the later 





Relation of 
surface to 

periods. The formula (and I hope that this 
is now clear) is involved in the artistic 
necessity of making the two visible dimen- 
sions expressive of the representational con- 
tent, so as to make the construction of the 
invisible third dimension 5 easy for us. 

This process of construction, of defining 
the limits within which the solfd -exists, is 
assisted and facilitated in Greek sculpture 
by the use of patterns and planes to serve 
as surfaces of reference, as it were spatial 
determinatives setting the limits within 
which our spatial construction of the solid 
must operate. 

(2) Planes of composition thus consti- 
tute a second device for facilitating our 
comprehension of spatial depth. The near- 
est point (to the eye of Jthe spectator) at 
which empty air ceases and statue begins 
marks a crucial distance, since it is at that 
point that our spatial construction of the 
work of art properly commences. If, how- 
ever, the statue-solid begins for us not with 
an isolated projection but with a considera- 
te surface all lying in one plane appre- 
ciably at right angles to the line of vision, 




the task of spatial construction is greatly 
simplified. And if, instead of a complicated 
system of varying distances from the eye 
the parts of the statue are ordered together 
into a series of planes appreciably parallel 
to the front plane, the task of referring 
every portion of the statue to its exact 
proper place in depth is now reduced to 
identifying merely the plane to which it 
belongs and to ordering the planes in their 
proper succession. Finally, if there is a 
similar indication of a back plane behind 
which no element of the statue extends or 
protrudes, the eye clearly grasps the spatial 
limits (in the dimension running away from 
the eye) between which the whole statue 
lies, and proceeds easily and surely to con- 
struct the statue-solid in its extension be- 
tween those limits. 

Now it is apparent that in the case of a 
modelled human body there cannot be a 
true flat frontal plane (such as we have 
described) if for no other reason than be- 
cause the body is everywhere rounded to a 
greater or less degree. But an effective 
and operative suggestion of a single plane 

A N D M O N O G R A P H S 



can be produced by patternising the lines 
and surfaces. It is characteristic of pat- 
tern that it pertains to areas which are seen 
together on an equal plane of projection. 
The pattern of a painting exists for us when 
we consider the painted surface as a flat 
area on which the various patches of colour 
lie; but the pattern vanishes insofar as 
we project the picture spatially in agree- 
ment with its representational content and 
read off these colour-patches as objects of 
a real world in spatial perspective (in fact, 
as a picture instead of a decorative design.) 
Insofar, then, as we visually hold together 
various surfaces and lines of a statue into a 
single pattern or geometric design which 
they make up, we tend to project them all 
upon a single plane at a uniform distance 
from the eye. 

Pattern can, therefore, force curvilinear 
surfaces into a single plane and thus assist 
in the process of breaking up the continuum 
of spatial extension in the difficult third 
dimension, into a succession of a few planes 
whose relative positions are more easily 



Pattern, it will be seen, works contrary 
to a three-dimensional apprehension be- 
cause everything to which it applies is by it 
marshalled into a single plane. Yet if it is 
not presented too insistently, pattern will 
serve to simplify the represenlata and 
thereby assist the process of visual appre- 
hension through which solid appearances 
are constructed. 

(3) But the most potent elements in 
this whole process are the numerous devices 
of linear suggestion by which the third 
dimension may be conjured up. These may 
be classed together and called "modelling- 
lines." When we make drawings in black- 
and-white, we may remove the look of flat- 
ness which attaches to surfaces between 
their outlines, either by appropriate sug- 
gestion of lights and shadows to suggest 
curvature or by adding properly curved 
interior lines. The Greek sculptor, adopting 
this latter device, carved lines whose cur- 
vature lay in the plane of the visible dimen- 
sion in order to suggest a curvature in the 
invisible dimension. This is a little more 
subtle than it sounds. Unlike a painting, a 



Relation of 
line to mass 



statue, being actually in the round, possess- 
es surfaces which are already correctly 
curved; but they may seem flat. The artist 
adds lines which do not follow the actual 
curvature of the projecting masses (since 
then they would no more seem curved than 
the whole surface in which they lie) but 
whose curve, lying in a plane more or less 
at right angles, cannot fail to be fully 
visible. He draws, in fact, a profile of 
the curved mass and spreads it out flat on 
that curved mass's own surface. 

If this is unclear to the reader, the 
nstance of an identical procedure in the 
decoration of architectural mouldings will 
be helpful Each of the mouldings of Greek 
architecture has its particular outline or 
profile as well as its proper surface-design 
or decoration. In their career along the 
side of a building these mouldings would 
tend to look flat. To obviate such an 
appearance, in certain conspicuous exam- 
ples the linear designs of the surface decora- 
ion are based on the curvature or profile 
of the moulding. Thus the egg-and-dart 
uses the curve of its profile to bound the 




little shields or "eggs" of its characteristic 

design; the Lesbian cymation employs the 

ogee of its profile for the outline of the pecu- 

liar leaves which are its characteristic deco- 

ration; the outline of the little beads of the 

string-courses echqes the profile of the 

moulding on which it is carved, and the 

rectangular-profiled fillet or taenia uses the 

rectangular-lined meander. The essential 

of the whole matter is this: these curves 

(which determine and outline the decora- 

tive design) lie in a plane normal to the 

ordinary line of sight, so that they are 

clearly visible and not foreshortened, but 

at right angles to the plane of curvature of 

the mouldings which they adorn. Their 

function is to suggest the existence of that 

curvature by putting its outline into the 

visible dimensions. 

This is just what is done in Greek sculp- 

ture, notably in the Pheidian perk/d. The 

"Fates" of the Parthenon pediment are 

outstanding examples of modelling (that 

is, projection in the invisible dimension) 

suggested by equivalent curves lying in the 





visible plane. The Venus Genetrix will 
serve for another example. 

Perhaps it is not too much to assert that 
the sculptor who has not divined this 
secret (however unconscious he may be of 
its exact formulation) cannot practise his 
art with the highest success. An attentive 
eye will recognise this device in much of the 
best work of the modern masters, in Michel- 
angelo very notably, but nowhere (I ven- 
ture to think) so consciously and consis- 
tently carried out as in the "great century" 
of Greek sculpture. 

The device of "modelling-lines" works 
directly contrary to the effects produced by 
patterns and planes. This, however, does 
not imply that these opposing formal 
methods are mutuaHy destructive and in- 
imical. Rather, the combination and just 
balance of their contrary influences in the 
same work of art constitute one of the 
remarkable opportunities of which a gifted 
sculptor will not hesitate to avail himself. 
The "Fates" of the Parthenon are again a 
most luminous instance and argument. 

Thus: (1) by intelligible pose with its 




expressive contours, and (2) by planes of 

composition with patterns to establish 

them, and (3) by the correct application of 

modelling-lines, the third dimension is 

shorn of much of its obscure content, with 

the result that we apprehend the statue- 

solid directly in its depth and spatial exten- 

sion. We apprehend it as a solid and not as 

a flat picture-like front-appearance behind 

which the invisible rest is assumed to lurk. 







Let it be granted that as a result of all 
these devices we actually do construct the 
solid spatially, that we do apprehend its 
depth directly and visually, in spite of the 
paradox that extension in the third dimen- 
sion is invisible 5 what then? 

It is here that physiological esthetics 
may find it convenient to break off; yet 
it is just here that they might become most 
valuable and interesting. Clearly the mere 
trick of apprehending a solid as a solid by 
means of the eyesight alone is very remark- 
able and entertaining, and may evoke the 
pleasure attendant on a heightening of our 
ordinary physical faculties; it may be the 
sine qua non of specifically sculptural emo- 
tion; but equally clearly it cannot be 
thought to be the sole and direct cause of 
that emotion. The visual apprehension of 
solids may be an indispensable factor in the 
sculptural emotion; but it is not in itself 
an adequate description or characterisation 
of that emotion. 

It seems clear that we cannot progress 



farther unless at this point we take into 
account the representational element the 
fact that the sculptor is not busy with a 
mere presentation of solids as solids, but 
that he embodies actual images of living 
beings and above all else delights in showing 
forth the human body. In Greek sculpture 
inanimate objects, except as the most 
minor accessories, were wholly ruled out 
during the fifth and fourth centuries. The 
lower animal forms occur in relief, but very 
seldom in sculpture in the round. Patently 
this was not mere accident or casual pre- 
ference, but betokened an appreciation that 
the proper sculptural study of mankind 
was man as though in some way the 
stereoscopic illusion obtained some secret 
value when it was concentrated upon an 
object which was after all essentially our- 

There is a present-day tendency among 
the initiate to consider the pictorial sub- 
ject-matter unimportant and irrelevant to 
the artistic values. This may be a good 
corrective for the persistent blindness with 
which a large part of the general public 



Relation of 
content to 



An impor- 
tant para- 
graph t 

fail to see anything in art except the 
represented. Taking the illusion as a 
reality, they proceed to interrogate their 
every-day feelings toward the objects under 
view, considering them as if they were real 
objects in a real world. This is particularly 
true in sculpture and leads to much naive 
embarrassment in the presence of the nude. 
Still, there can be no doubt that the repre- 
sentational element is actually of the 
greatest import to sculpture. 

The direct visual apprehension of a solid 
as a solid (which we have defined as the 
prime esthetic differentia of sculptural con- 
templation) is psychologically relevant but 
artistically unimportant unless the solid is 
also apprehended as an embodiment of an 
animate object. The sculptor's device is 
a means of giving us the most direct and 
intimate apprehension of the animate and 
animating qualities and forces of the object 
whose appearance he has imitated. 

Whether or not we project ourselves 
sympathetically into the statue, I do not 
know. I feel that the cardinal objection 
to the empathy theories is their failure to 




measure up to our honest and actual exper- 
iences. They are partly true, certainly; 
but they make up only a small part of the 
mystery. They exaggerate a contributory 
factor into a paramount issue. Still, there 
is some very clear connection between our 
apprehension of the solid as solid in the 
case of a human statue and the fact that 
the only other method of direct apprehen- 
sion of spatial solidity is our awareness of 
it for our own body. It may not be strictly 
our own strength or power of movement or 
agility or gravitational equilibrium which 
we feel to be heightened while we contem- 
plate the sculptural indication of these 
qualities; but our power of spatially appre- 
hending the statue as we spatially appre- 
hend our own bodily selves makes us 
immensely susceptible and sympathetic to 
an emotional and almost muscular-physical 
understanding of such qualities when they 
are sculpturally presented. We know them, 
not from the outside (as we comprehend 
strength when we see a strong man pull or 
push or lift, as we comprehend speed when 
we watch runners, or agility as when we see 




First con- 
from our 
(Cf. chap.I 

a mountain-goat) but from the inside, in 
terms of what they are for their possessor 
And we apprehend them without strain or 
discomfort or other physical preoccupation 
They come to us direct, simple, intense 
sheerly pleasurable. Herein may be the 
ultimate secret of the sculptural appeal 
It is, therefore, not enough that we appre- 
hend a solid as a solid, nor yet that we com- 
prehend that the solid is (let us say) a 
human being. More than that, we must 
apprehend the animate and animating 
forces and qualities. We must apprehend 
them not by an intellectual inference nor 
by a mere second-hand sympathetic under- 
standing; in sculptural contemplation we 
apprehend them still more immediately, we 
eel them directly and as it were from 
within. The mechanism by which this is 
accomplished is purely formal. It consists 
n devices of line, surface, and gravitational 
arrangement, to which the represented 
object is made to conform. These devices 
are in themselves abstract, geometric, and 
ittle evocative of emotion; they assert 
heir magic only when they are correctly 




fused with representational matter. So 
used, they are all-important; indeed, the 
proper employment of these pure forms is 
the second great secret of the sculptor's art 
At the risk of repetition, these pure forms 
of sculpture must be here enumerated 
They include all the devices of one-dimen- 
sional emphasis which were discussed 
in connection with linear presentation. 
Through these are occasioned various 
suggestions whose common characteristic 
is the absence of all suggestion of gravita- 
tional weight or mass. Then there are the 
devices of two-dimensional emphasis which 
were discussed in connection with relief 
carving. Of these the most notable is pat- 
tern; but there are numerous other ones, 
all dealing with the arrangement of areas. 
Through these are occasioned suggestions 
of size, internal structure, synthesis of 
forces, equilibrium, and other similar mat- 
ters, of which the common characteristic is 
their suggestion of static mechanics with- 
out much suggestion of motion or of mater- 
al solidity and weight. Finally there are 
the devices of three-dimensional emphasis, 





and with these we have already dealt at 
length. It is the fusion of the suggestions 
accruing from an artistically satisfactory 
use of all these formal devices which raises 

sculpture out of the category of a merely 
imitative craft which faithfully copies in 
terra cotta, stone, or bronze, the shapes and 
appearances of men, women, and animals, 
and makes of it an art with all an art's 

equipment for 

arousing great human 





If it is true that the sculptural art bor- 
rows various devices peculiarly appropriate 
to certain of its sister arts, we can perhaps 
find therein one reason why artists are so 
requently impressed with a sense of the 
community of all the arts, even though 
they seldom manage to bring this idea to 
the point where they can make clear what 
it is which they think the arts have in com- 
mon. In order to do so it is first necessary 
to recognise the extreme complexity and 
variety of the esthetic emotions occasioned 
by any particular art and to sort out as 
many contributory factors as possible 
One would then discover that certain oi 
these factors recur in the esthetic complex 
of emotions incident to other arts. If we 
have once come to recognise that nearly 
every kind of stimulus derivable from 
almost anything in heaven or earth may be 
associated in the state of a cultivated mine 
under the impression of poetical suggestion 
we will not feel that there is anything un 
usual if the arts borrow various ranges o: 



actors in 
he sculp- 
ural appeal 



emotion from one another. It would be 
beside the point to make this analysis for 
all art; but on the basis of our previous 
discussion it should be quite easy to make 
this analysis for sculpture. It would read 
somewhat as follows : 

The human body in sculptural represen- 
tation differs from its prototype, the living 
body, by calling attention to certain quali- 
ties which otherwise tend to pass unnoticed. 
By being motionless it presses on our atten- 
tion its particular volume and configura- 
tion. This immediate apprehension of the 
mass or volume of a represented object is 
the primary differentia of the sculptural 
appeal. This apprehension of a three- 
dimensional solid in terms of its surface 
aspect and its bounding-lines clearly recurs 
in the contemplation of architecture. In 
utilising this, architecture accordingly bor- 
rows from the sculptural appeal. Our 
modern architects are , not sufficiently 
aware of this. They are reliant upon paper 
plans and paper elevations; they design 
with one or two contours in mind, that is 
to say, those which they set down in these 



paper elevations. But when their building 
is built in three-dimensional matter, it will 
reveal a host of contours which these archi- 
tects never considered; it will in fact be 
subject to all the devices which occasion 
the characteristic appeal of the sister-art 
of sculpture. Would not greater sensitive- 
ness to this sculptural element save us from 
some of the paper-flat facades, unfeeling 
sky-lines, and brutalised corners with which 
our modern architects present us; or are the 
exigencies of rectangular building lots, 
retailed at fabulous prices, fatally opposed 
to such an artistic renascence? That all 
depends, in the long run, upon what We 
really want humanism or industrialism, 
the highest spiritual attainment or the great- 
est bodily comfort of the greatest numbers. 
In addition to this characteristic sculp- 
tural factor there are always other factors 
prominent in sculptural art. We cannot see 
a statue without seeing its bounding-lines 
or contour; and as we move around the 
statue this contour continuously goes over 
into a different one, so that our emotional 
apprehension of the outlines is not merely 





equivalent to an almost numberless series 
of relief appearances viewed from points 
on the circle, but also includes the transition 
of one aspect into the next and the mental 
reconstruction of all these transitions in 
agreement with the shape of the three- 
dimensional mass which we have all the 
time been apprehending. There is thus a 
very powerful and inseparable fusion of a 
linear element with the mass element, and 
these two are in mutually complementary 

Thirdly, there is in sculpture the factor 
of chiaroscuro or of light and shadow, inti- 
mately involved in that appreciation of 
projections and hollows which determines 
our apprehension of mass. But the distri- 
bution of light and dark constitutes an 
arrangement of differently illuminated 
surfaces whose mutual relations of size, 
shape, and position introduce esthetic fac- 
tors that are prevalent in the art of painting 
and drawing and which may be called pic- 
torial. Since every surface has a bounding- 
line and may be diversified with interior 



line, a further linear element has been intro- 

Lastly, all except very small statues 
impress us as having weight; and even if 
we fail to be aware of the actual weight of 
stone or metal and consider only the repre- 
sented object, we shall have a sense of 
weight and gravitational pull overcome in 
that equilibrium of forces which is the secret 
of the statue's balance and immobility. 
This sense of weight and thrust and balance 
is so prominent and characteristic a factor 
of the architectural appeal, that we may 
say that here sculpture borrows from arch- 

Clearly the importance of these various 
components or factors may vary enor- 
mously. In a figurine the sense of weight 
and gravitational pull is at a minimum; 
in a colossus its strength is enormously 
enhanced so that it may become one of the 
most prominent factors. The handling of 
grooved troughs and hollows and under- 
cuttings will determine the range of dark 
and light, as will the depth to which light 
penetrates the material before it is wholly 





absorbed or reflected. Some sculpture is 
much more linear than other. Gothic has 

far more abundant shadow than Greek: 

Greek in general has more insistent line 
than Gothic. And these are but the most 

obvious and powerful components in a 
result which is after all as complex and as 
diverse as are the works of sculpture which 
men have made or the minds of men which 

have contemplated them. 

After this survey of the principles and 
practice of Greek sculpture, it should be 
possible to understand somewhat more 
clearly the nature of the sculptural appeal 
as it comes to us from fifth and fourth 

century Hellenic art. 
We have seen that the fundamental 

device of the art is the visual presentation 
of a solid so that it may be directly and 
immediately apprehended in all its spatial 
extent and depth; that because of the 
immediate and sensuous character of this 

apprehension, we are able to sense strongly 
and clearly the play of forces which ani- 
mate this solid, when we see it not as mere 





marble or bronze but as that which it repre- 
sents, an animate being like ourselves 
(though perhaps a recognisably more har- 
monious and perfect being, devoid of the 
imperfections of material evils) and that 
finally certain purely formal devices of the 
art facilitate and amplify this apprehension 
of animating forces, which in turn reacts 
upon our own sense of animate energy. Out 
of this emotional apprehension, occasioned 
in this manner, arise all those feelings of 
physical animation, elation, exaltation, or 
whatever vague and inexpressive words 
we may use in trying to convey to others 
verbally a sense of our innermost experience 
from an artistically effective piece of 

Let it not be thought, however, that the 
foregoing is intended as an exhaustive 
account of the emotional content of the 
sculptural appeal. Purposely it has con- 
fined itself to the most distinctive and 
characteristic elements. But with these 
are blended amazingly many minor fac- 
tors: our associations of every -day life 
which are called up at sight of the repre- 

Still other 




sented object, would alone be a highly com- 

plex element, especially when we remember 

the range from mere sensuous delight to 

direct sexual stimulation that may be pre- 

sent in the contemplation of the nude. It 

would be foolish to assert that these asso- 

ciations are in no way present or involved. 

But apart from all such artistically extran- 

eous elements (extraneous, since they reside 

more in the representational illusion than 

in the actual working of the art as an art) 

and amid all this characteristically human 

complication of elements, it ought to be 

possible to trace the dominant and peculiar 

emotional contribution which the exigen- 

cies and possibilities of the sculptor's art 

occasion ; and this I have tried to do inso- 

far as the conventions and underlying prin- 

ciples of Greek art may be accessible to our 

analysis to-day, two thousand years after 

the disappearance of the civilisation in 

which that art was so conspicuously and 

successfully practiced. 




Quite apart from all these devices of pure 
form, the representational element in 
Greek sculpture was modified in the inter- 
est of that quality called Idealism the 
property of Greek art which more than 
any other has been extolled and misunder- 
stood by succeeding centuries of classical 
enthusiasts. Into the true character of 
this "idealism" in Greek sculptural repre- 
sentation of the human form it is essential 
that we should now enter. In order to do 
so, it will be useful to discover first how these 
idealising tendencies arose, since they are, 
in their origin, almost accidental inherit- 
ances from the primitive and archaic 

Truth to Nature, fidelity to the actual 
shapes and appearances, is the lodestar that 
leads the sculptors on. It is also the song 
of the sirens, a Lorelei to bring their ships 
upon the rocks ; for in the very moment of 
attainment the true art of sculpture goes 
to wreck and vanishes from sight, broken 
on the reefs of a mere imitative realism. 


The repre- 
Truth to 

always a 
end or aim 




the devel- 
opment of 

its periods, 
viz., (1) 
(2) archaic 

That this is really so, may be easily (but 
not briefly) proven by an examination of 
the history of any great epoch of sculptural 
attainment, though we must consider both 
the technical history and also the esthetic 
properties of sculptural form. For my 
particular field of Greek art, I must assume 
that the historical development is suffi- 
ciently known to permit a very brief sketch 
of it in place of a chapter-long chronicle. 

Greek sculpture in its beginnings. was 
entirely primitive. In the course of the 
sixth century B. C., it passed through its 
archaic period, a period which reached at 
the beginning of the fifth century a high 
degree of stylistic development and yet 
remained clearly "archaic" in quality. 
This quality is very clearly and accurately 
definable, and depends primarily on the 
substitution of set schematic forms for the 
irregular and variegated appearances of the 
real world. An artist of this period cannot 
master all the torsions and tensions, the 
projections and cavities which real objects 
present, but confines himself to a simple 
standard appearance from a single and 



favorable point of view. The long locks of 
hair which fall over the back and shoulders 
present in life a great complexity of disap- 
pearing and reappearing lines and surfaces ; 
but the archaic sculptor devises a flowing 
zig-zag groove to suggest the general look 
of these long locks, and repeats this in 
parallel succession over the surface of stone 
which he has blocked out for the hair. 
Neatness of execution and sharpness of 
line become the obvious criteria of success, 
so that the great archaic craftsmen will 
carve these zig-zags in the most mar- 
vellous abundance and the most finished 
delicacy. But no matter how much they 
labor, they will not bring the stone to the 
appearance of actual hair, because the 
schema of the parallel zig-zag is only a 
geometric and conventionalised equivalent 
for the true surfaces and projections of the 
living model. 

A different schematic design will be 
devised for the curled locks over the fore- 
head. Schemata will be found for eye- 
socket and eye-lid, ear, mouth, muscles of 
torso and back, fingers, toes, and all the 





rest of the body. And all these schemata 
will be characterised by a geometric sim- 
plicity and an unvarying repetition. 

Consequently, as the archaic period ad- 
vances, a wonderful decorative abundance 
of rather abstract linear forms will cover 
the sculptural surfaces more and more pro- 
fusely till they seem as rich as tapestries 
or oriental rugs. This archaic craftsman- 
ship, full of the most loving detail and the 
most engaging intricacies, makes a very 
strong appeal to the modern world, which 
has no difficulty in comprehending its 
devices and desires. 

In Italian painting Botticelli stands 
near the culmination and close of the 
archaic period of his particular epoch of 
painting. One has only to look at the 
linear forms for hair, drapery, features, 
waves, flowers, and shore-line in the famous 
Birth of Venus, to discover over again all 
the schematic devices of late sixth century 

Gothic sculpture in the twelfth century 
A. D. was passing through the same phase. 
So similar are its schematic devices that 




many an untrained observer will scarcely 
be able to tell you, in the case of two photo- 
graphs of sculptured heads,, which is archaic 
Greek and which is archaic Gothic. 

In all these periods archaism dies the 
same death . I ts final over-refinement forces 
a realisation that it is flagrantly and funda- 
mentally untrue to actual appearances. A 
period of simplicity ensues. The repeti- 
tions, the parallelisms, the minute scale 
for surface-lines, are abandoned. A few 
forms are used, but these are as like the 
real thing as the sculptor's awakened sense 
for realism can achieve. 

The bronze charioteer of Delphi will 
serve admirably for illustration. Its sim- 
plicity is largely due to the preservation of 
the schematic devices of the archaic period, 
which are kept unchanged in their essential 
shape. The long vertical flutings of the 
drapery are unbroken in their career of 
alternate light and shadow; yet their pro- 
files depart subtly yet sensibly from the 
set geometry of their implicit schema. There 
is almost the formality of a fluted column; 
yet, unlike such a column, the arris against 

(3) transi- 




the sky flickers and wavers, and the outline 
of the driver runs through a whole series of 
changing contours as we move around the 
statue. For the sleeves of the garment a 
more finely wrinkled schema is set, every- 
where repeated, and everywhere tempered 
from strict parallelism and geometric repe- 
tition by tiny irregularities and departures 
from the norm. With the schematic forms 
for the hair, with the set geometry of curves 
for eye-brows and eye-lids, forehead and 
cheek and chin, it is alwajrs the same 
the uneven tremor of life shaking the un- 
physical perfection of the mental construc- 
tions of an ideal geometry. 

In all this, one point must be labored at 
risk of over-emphasis. To each element of 
the representation (free-hanging gown, 
close-drawn drapery, hair, features) there is 
allotted only one standard form, a schema 
based on the geometric simplification of the 
actual appearance, in conformity with its 
visual memory-image. This single form 
must be repeated as much as necessary in 
order to cover and inform the surface to 
which it applies. Thus the schema for 




free-hanging drapery must be repeated over 

all the drapery which hangs free. There 

results a clear division of the whole statue 

into areas each characterised by some pecu- 

liar line-form or pattern which everywhere 

distinguishes it and which is inherently 

appropriate to the representational inten- 

tions at that point. That is why, in the art 

of this period, critics find no "irrelevant 

lines" (how should there be; where would 

they come from?) and are complimentary 

toward "structural simplicity" and clear 

structural articulation. For the Greek these 

high effects were no mystery. He was 

merely keeping to his inheritance of sche- 

matic forms and tempering them in the 

direction of the confused diversity of the 

real world. In consequence . of his own 

method, he could not fail to have his sensi- 

bilities awakened so as to grasp the secret 

of large construction through simple homo- 

geneous areas and of simplicity of effect by 

correlation of similar linear forms. The 

accidents of primitive draughtsmanship 

were beginning to go over into esthetic 





(4) strong 

devices amenable to the imaginative voli- 
tion of artistic creation. 

Certainly, the archaic schemata are 
subjected to improvement so that their 
fidelity to Nature is being constantly aug- 
mented. And since the correct rendering 
of living shapes is, after all, the supreme 
technical objective, we should expect .to 
see a very uneven attainment and even in 
the same piece of sculpture find the tradi- 
tional renderings side by side with unex- 
pected vivid and brilliant bits of natural- 
ism. This is precisely the quality which 
frequently occurs to puzzle the' attentive 
student of the period. The pedimental 
sculpture of the Zeus temple at Olympia, 
the Boston counterpart of the Ludovisi 
throne, the Esquiline Venus (if she be 
rightly classed as of this period) reveal just 
this uneven command over true appear- 
ances. But everywhere the framework of 
the older tradition is apparent. 

When the constant repetition of a set 
linear schema over an entire plastic area 
palls on the sculptor (just as the uncritical 
decorative parallelism of the archaic period 




had palled on his predecessors) the transi- 
tion period gives place to the period of 
strength, the age of the great masters of 
the Periclean period, the Athenian Pheidias 
and the Argive or Sikyonian Polykleitos. 
Naturally no break with the preceding 
generation is discernible. The schematic 
tradition still lurks under the linear pre- 
sentations, and the very same forces which 
gave to the Delphic charioteer its harmony 
and simple greatness are still operative. 

It follows that the work of the Pheidian 
period shows only an incomplete approxi- 
mation to naturalism. In the preceding 
period truth to Nature tempered the sche- 
matic forms; now the schematic (and other 
more 'purely artistic) forms temper truth 
to Nature. Yet a complete truth to Nature 
can be attained only by a far-reaching and 
accurate grasp of detail, and this is too 
difficult to be achieved immediately; only 
the broader and more essential elements 
are recorded. There results a style of great 
breadth and power: the archaic has gone 
over into the strong period. The sculptor's 
attention is inevitably drawn to the larger 




(5) fine 

formal effects of broad surfaces and con- 
spicuous lines instead of to the multiplica- 
tion of decorative detail. He learns what 
effects come from the balanced masses of 
unbroken surfaces and from the travel of an 
uninterrupted line, and his work gains 
immeasurably in power by the suppression 
of "irrelevant detail" which is forced upon 
him by his inability to fashion those details 
and to include them. 

Pheidias and Polykleitos are the great 
masters of this period. The outstanding 
elements of their style are precisely those 
of the period in general. So, Michelangelo 
is the "strong" sculptor of the Renaissance. 
He could scarcely have been so had not his 
lifetime happened to fall in the strong 
period of his sculptural epoch. 

Inevitably, in Greek sculpture as in cor- 
responding other times, as the period 
advances, greater and greater approxima- 
tion to natural truth ensues, detail upon 
detail is correctly apprehended and em- 
bodied in the sculptural corpus. There 
comes a time when the more purely formal 
traditions of lines and masses, though they 




are still operative, are nearly obscured by 
these formally irrelevant details which 
stricter fidelity to Nature has introduced. 
Just at this stage and because there is so 
nice a balance between the illusion of 
objective appearance and the more geo- 
metric and abstract beauty of sculptural 
form, there arises a period of very great 
distinction. To most later-day collectors 
(and to all that period's contemporaries) 
this will seem the period of finest flower. 
It will, indeed, be the period not of strength 
but of beauty. It is the period of Praxi- 
teles, of full-blown Gothic, of Raffael and 
Titian, of Benvenuto Cellini with his 
Perseus and Medusa, and the youth of 
Bernini when he did his group of Apollo 
and Daphne. 

But the same force which brought this 
period on also brings its change and disap- 
pearance. The realistic details increase 
until fidelity to Nature makes most of the 
curlier formalisations impossible. The 
artists copy the human body as it is: so 
much the worse if it has not the formal ele- 
ments of power and beauty upon it. The 

(6) free or 




artist must find his expression through 
things as they really are. The stone or 
bronze is almost a replica now of the living 
model. The model therefore, like an actor, 
must itself embody the emotional expres- 
siveness which the artist intends to com- 
municate to his public. Statuary is almost 
the craft of turning cleverly posed tableaux 
vivants into marble or metal. Human 
bodies are chosen as models for their own 
natural beauty, as in the Syracuse Aphro- 
dite or the Capitoline Venus (or even for 
the stimulation from their ugliness or unus- 
ual qualities, as in the Gauls of the Perga- 
mene artists); poses are reproduced liter- 
ally for their physical attraction or their 
energy, strength and power displayed in 
aettial use. This last period is accordingly 
complex and many-sided. It is not merely 
the period of Naturalism and Expressive 
Realism, but of Theatricalism and Sensa- 
tionalism wherein (since everything must 
proceed from true and faithful appearances) 
the objects of the real world are tortured 
into striking and emotionally moving poses 
or combined in exciting situations. We 



have the Drunken Woman, the Laokoon, 
the Farnese Bull, the Borghese Warrior, 
and all the other familiar virtuosities of the 
Hellenistic Age. 

With the culmination of this period 
comes the attainment of that goal which 
glimmered above the crude beginners of 
the art and led them to find geometrically 
simplified equivalents for the bewildering 
unseizable variety of seen appearances, 
and thereby opened up the archaic period. 
This same glimmering and unreachable 
objective destroyed the artificialities of the 
archaic, produced the simple forms of the 
strong period, gave the wealth of detail 
which turned the strong into the fine, and 
at last overwhelmed the period of beauty 
with the abundance of naturalistic detail. 
Throughout, the same power makes each 
period and destroys it, and at the last so 
dominates the sculptor's art as to leave no 
room for the other elements from which 
it drew most of its power and appeal. At 
the end, when the goal is attained and no 
further direction of progress is apparent, 
the sculptors, aware of the aridity of their 


(7) eclectic 




, I 



periods a 

cycle in the 
epochs of 

success; turn mournful and wistful glances 
back toward the earlier periods whose naive- 
delight and unsophisticated energy and 
enthusiasm are denied to their accomplish 
cd and facile days. Most of all, the period 
whose contrast with their own is most 
flagrant, the period of ripe or even over- 
full archaism, attracts them. There/ensue 
"revivals," precious enthusiasms of the 
dilettanti the fads of the Nc -at ticists. 
the enthusiasm of the Augustan Age for 
Kanachos and Kalamis, in more modern 
times the neo-Gothic, neo-Grcek, and all 
the other resuscitations which have made 
of modern architecture an eclectic thing of 
copy-book and pastiche, the Prc-Raphael- 
ites in painting, the present-day enthusiasm 
for the archaic Greek and all undeveloped 
periods of artistic nascence the world over. 
Sub specie acternitaiis, artistic epochs are 
cycles, each moving through identically the 
same phases under the same forces to the 
same ultimate result. What most aston- 
ishes the student of these cycles is the 
inevitable progress of each phase to its 
own destruction, all because an artistically 



almost irrelevant objective (fidelity to seen 
appearances) exerts a supreme control, 
while the seemingly true and relevant 
objective of the art (control of the esthetic 
forms which determine the sculptural 
appeal) is never clearly set up as the true 
one to be striven for. Rather than the 
question "What are the expressive forms 
of my art, into obedience with which I 
must mould my matter?" the sculptor 
seems to put himself the query "Have 1 
caught the true look and fashion of the 
things which I portray?" Apparently, the 
one criticism which artists have been least 
able to endure or to meet, has been the 
artistically rather gratuitous one "You 
have failed to copy Nature correctly!" 

Removed by lapse of centuries from the 
whole cycle of Greek sculpture, we can see 
that the works of each period have artistic 
value and that what seemed at the time to 
be progress, rendering one's predecessor's 
methods obsolete, was always realistic 
advance but not necessarily greater artistic 
achievement. How we are to rate these 
various periods for this, their artistic 





Relation of 
fidelity to 
forms, in 
the (1) 

achievement, is a problem for the Canon 
of Taste to decide. Some critics will see 
the peak already attained in the late 
archaic, and class all subsequent products 
as decline; others prefer the period of 
strength, others the period of beauty; few 
will put their preference so late as the final 
stage of Naturalism or Eclecticism. 

The present study, being an inquiry into 
artistic forms, must leave unraised the 
questions Which belong to the province of 
Taste, and. without passing judgement on 
relative merits, must try to discover how 
the pure forms of sculpture were employed. 

For the archaic period these "forms" 
have been abundantly studied; but it is 
essential to remember that they are in their 
origin artistic accidentals, mere devices of 
convenience by which the early artist helps 
himself out of the difficulties of imitating 
seen appearances, and that they only 
acquire artistic significance in proportion 
as the sculptors learn to turn them to 
account for purely artistic rather than 
merely representational ends. There is 
here a very present danger of confusing 



Origin with Validity (as the manuals of 
logic sometimes term this fallacy) and of 
imagining that because we have discovered 
how the various conventions and devices 
arose we have therefore fully explained and 
understood them. 

In order to throw light on the peculiar 
conventions of archaic art, the behaviour 
of savages and children has been much 
interrogated. The purpose is by wide- 
ranging comparisons to elicit the presence 
of certain common conventions in all primi- 
tive art and to explain these conventions 
on general psychological or material 
grounds. The results have justified the 
attempt and have led to a wholly new un- 
derstanding of artistic origins in Greece and 

Briefly, these comparative studies have 
established the thesis that the primitive 
draughtsman does not draw from life but 
from memory. Those of us who have no 
knack for drawing may perhaps appreciate 
the unreasonableness in a demand that we 
should "just draw thipgs as they look; put 
down what you see!"fand realise that until 




a race has begun to learn to draw, the actual 
changeful appearances of things elude any 
attempt to find their direct and faithful 
linear equivalent. Primitive drawing is 
not so much a crude and incomplete ren- 
dering of the correct outline, as a sort of 
visual embodiment of those notions for 
which nouns and names are symbols. 

We can perform the crucial experiment 
ourselves. At each of the names, "hut," 
"frog," "fly," a mental image is called up 
with more or less distinctness. If we try 
to transfer this mental picture to paper, we 
shall discover that it possesses some very 
surprising and yet perfectly reasonable 
characteristics. Since it is only a general 
image applicable to any of its species, it 
will show only essentials, but it will include 
all these. A quadruped must have four 
legs, a bird must show two wings; a human 
face will appear in full-front view or per- 
haps in profile with full-front eyes. In 
fact, a primitive drawing tends to be a 
mental construction exhibiting just those 
observations and analyses which the mind 
has made for its own pragmatic interests. 




Accordingly, we should not be surprised at 
finding in a primitive picture or statue of a 
human being an almost logical rigour. The 
right half of the body will have no reason 
to differ in position or appearance from the 
left half ; the head and eyes will turn neither 
to this side nor to that; the arms and hands 
will hang close to the sides or be raised 
evenly or be folded symmetrically. One 
lock of hair will look exactly like another, 
unless the memory of the artist has record- 
ed the different appearance of the short 
hair on the forehead and crown and the 
long back-hair, in which case there will be 
two hair-forms in use. Inevitably, then, 
certain set forms or schemata will be 
employed for the various parts for head 
and face, with hair, eyes, nose, and mouth; 
for fingers, navel, chest, ribs, and so on. 

How will primitive art advance, how will 
it improve upon this initial stage? Not by 
discarding these memory-images and sche- 
matic forms, but by perfecting them so 
that they are more faithful to actual 
appearances, and by adding supplementary 
schemata for details which have been over- 




looked and therefore omitted. At each 
repetition of the standard theme of the 
erect nude male in early Greek sculpture 
you may notice some trivial improvement 
of the old or addition of the hitherto unob- 
served and new. But, however much this 
progress is continued, there cannot be 
more than a specious approximation to the 
photographically true and exact. Between 
the eye which sees ai\d the hand which 
gives back the seen, there intervenes a 
something which never lies between the 
camera lens and the photographic plate: 
I mean the mind, to which the eye reports 
and from which the hand takes its instruc- 

It is the charm of archaic art that it 
transmutes the vagaries of the external 
shapes of our ordinary world into the 
almost intellectual precision of these men- 
tal constructions and, by lavishing a most 
effective diligence on the perfect execution 
of each part, makes only the more apparent 
the mental conversion through which the 
natural appearances have been made to 
pass. The more it labors for perfection, 




the more clearly it calls attention to its 
departures from objective truth, because 
archaic art can only repeat the generalised 
form (however minutely and succinctly) 
while Nature never shows this underlying 
type, but only the thousand and one indi- 
vidual variations of it. 

Little by little this secret dawns upon the 
generations of artists. They discover that 
truth to Nature lies not in set repetition 
but in indefinite variation. The great 
problem (and the profound verity) of "the 
One and the Many" lifts upon their hori- 
zon. The form is one, but its manifested 
appearances are many. There is one form 
for the oak-leaf, another for the cypress; 
but when Nature makes an oak-tree, she 
does not hang it with a thousand identical 
leaves. Every leaf is different; yet every 
leaf exhibits the same form, which is the 
form for oak-leaves. In sculpture, the 
schematic designs for hair and for drapery- 
folds are retained as the basic structure 
from which every wisp of hair, every fold 
of drapery, is allowed to depart by a slight 

(2) transi- 




(3) strong 

This is the period of transition from 
archaism to freedom, and the sculpture 
which it produces shows a restrained and 
quietly severe beauty, as free from the 
affectations of merely decorative ornament 
as from the inconsequent lines and inten- 
tionless diversity of naturalistic imitation. 
In Greece this phase occupied roughly the 
first half of the fifth century B.C. 

Because fifth century Greek sculpture 
inherited all the schematic forms for repre- 
senting objects, it could not be true to life. 
At the time of Pheidias and Polykleitos, 
the sculptured human body was put 
together from a series of parts each of which 
had a more or less intellectualised shape or 
structure, and each of which had a conse- 
quent bias toward geometric formalism 
and geometric simplification. The idealism, 
the "classic restraint," the omission of non- 
essentials, which are so generally acclaimed 
to be an outstanding mark of Greek sculp- 
ture, thus had their origin. Canova and 
Thorwaldsen and other neo-classicists 
worked by elimination, by deliberately try- 
ing to suppress complete natural fidelity: 




Polykleitos was still adding detail upon 
detail to his heritage and was striving to 
give to his inherited forms that freshness 
and vigour which only the desire to copy 
Nature can impart. They were going in 
opposite directions, the old Greeks and the 
modern neo-Greeks; and since we must 
judge them by their aims and ideals, we 
must allow that they are poles asunder in 
what they managed to accomplish. 

The most lucid commentary on the art 
of Polykleitos is to be found in the lyric 
poetry of Bacchylides and the dramatic 
poetry of Sophocles. Both these masters 
worked with forms inherited out of the 
archaic period of their art; and both found 
their highest artistic opportunity in that 
minute and subtle modulation of the un- 
realities of formalism which would approxi- 
mate more closely the real world about 
which (after all) they were writing. This, 
as far as I can understand it, is the typical 
mid-fifth-century mentality. It is only 
possible in a period where the public is 
familiar with all the artistic forms and con- 
ventions with which early art is trammeled, 




and so can appreciate the minute temper- 
ings of those forms by which the masters 
get their effects. On this condition 
depends the tnie Sophoclean "irony." And 
only if we bear this condition in mind, can 
we appreciate the full force of the saying 
attributed to Polykleitos by Plutarch 
Tolykleitos the sculptor said that the work 
was most difficult when the clay was on the 
nail." However we may explain the 
exact meaning of the phrase iv ow^t (ad 
unguinem) it is clear that Polykleitos was 
referring to the importance of minutiae 
for the effect of the whole. Modern critics 
have often sighed enviously at the mar- 
vellous finesse of appreciation Which they 
ascribe to the ancient Athenian audiences 
which sat in the Theatre of Dionysos. But 
this audience was able to appreciate 
Sophocles because they had grown up in a 
knowledge of the forms and conventions 
by minute deviation from which the 
Sophoclean drama gets all its tang and 
subtlety. It takes a similar training to 
appreciate Polykleitos. 




We said that in the time of Polykleitos 
the sculptured human body was put to- 
gether from a series of parts each of which 
had a more or less intellectualised or geo- 
metricised shape and structure. Out of 
these the sculptor put his figure together 
somewhat as the architect built up his 
Orders out of their component members. 
Now in architecture it was patent that 
the relative shapes and sizes of these mem- 
bers determined the fitness and Tightness 
of the whole Order. Since the parts of an 
architectural Order are man-devised and 
not copied to imitate living things, the 
architect can determine their sizes to suit 
himself or rather to suit those peculiar 
demands of fitness and rightness. But if 
fitness and rightness are a function of 
relative shapes and sizes, they must be 
produced by ratios, by number. Number, 
at work in all the elements, engenders the 
perfection which is their beauty. 

There was, in the fifth century, an 
obvious parallel between the architectural 
Order and the sculptural Order of the 
human statue. This, too, was put together 

peculiar to 

his period 




out of parts each of a standard shape. Here, 

- . 

too, the fitness and Tightness of the whole 

somehow depended on the relative sizes of 

these parts. -What more natural than to 

draw the same conclusion and believe that 

Number here too was sovereign? Number 

is certainly sovereign in Nature to a most 

amazing extent (far more than we ordi- 

narily dream, until we pick up a pamphlet 

on phyllotaxis or shell-formation or pro- 

portional growth), and the Greek may well 

have known this, especially as the Pytha- 

goreans seem to have taught specifically 

such a doctrine. In any case there is little 

doubt that Number was introduced into 

sculpture as it had been into architecture 

and, probably, into the vase shapes of 

pottery. 7 

The locus classicus is in Galen, where 

the following passage occurs: 

Chrysippos holds beauty to consist 

in the proportions not of the elements 

but of the parts, that is to say, of finger 

to finger and of all the fingers to the palm 

and wrist, and of these to the forearm, 




and of the forearm to the upper arm, and 
of all the parts to each other, as they are 
set forth in the Canon of Polykleitos.* 

Here we have a clear statement, from a 
somewhat late and round-about source, but 
expressed in just the language we should 
have anticipated. Commensurability of 
parts (</A//,T/o6a) is the working of Number 
in sculpture and the earnest of its perfec- 
tion. There is an almost metaphysical 
belief that beauty and the ideal type for 
sculptural representation are characterised 
by an almost supersensual. because intel- 
lectual and mathematical, structure. 

This is not a unique or unusual turn for 
the artistic mind to take. It is not uncom- 
mon for artists to seek refuge in formulae 
and to look for a sanction for sensuous artis- 
tic beauty in canons and pseudo-scientific 
abstractions. For example, in a period 
when geometric science and art were both 
in renascence, Albrecht Durer of Nurnberg 
was obsessed to discover the underlying 
geometric and numerical necessity in the 
beauty of the human form, and he succeed- 




ed in producing a canon of perfection which 
was very full of Number. But he came to 
see that this canon was a standard from 
which to deviate, not a standard toward 
which to strive. This is just what we learn 
when we study the role of Number in the 
beauty of natural objects such as flowers, 
leaf-forms, and shell-forms. The beauty 
of living things seems to involve a host of 
minute departures from an underlying 
geometric uniformity. The Canon is this 
underlying uniformity, the eidos of Greek 
philosophic speculation. It is made up of 
many numbers; but the mere literal incor- 
poration of these numbers in an artistic 
representation will fall short of beauty 
because it will show only the mathematics 
and not the living thing. This seems to be 
the meaning which Polykleitos intended to 
convey when he said that the work was most 
difficult "when the clay was on the nail." 
I venture the opinion that Polykleitos 
meant that it was not hard to follow the 
inherited schematic shapes and standard- 
ised proportions, but that the real art lay 
in the minute departures from the stand- 



ard norm, which converted the statue 
from a geometric manikin into a sculptural 
equivalent for living reality. It was in 
making these little touches that the clay 
was "on the nail." 

Accordingly, when we take actual 
measurements of these fifth century canon- 
ic statues or their replicas, we signally fail 
to detect those simple arithmetical ratios 
which we must imagine the sculptors to 
have incorporated. This very much vexed 
and discountenanced an earlier generation 
of archaeologists. Yet the reason is not far 
to seek. The forms of architecture, being 
man-devised and conventional, owe no 
resemblance to natural objects, so that 
their proportions can be altered to suit 
man's fancy and esthetic sensibility; but 
only a slight alteration in the relative meas- 
urements of the human organs will overstep 
the limits imposed by sculpture's express 
aim and intention of representing under 
plastic form types of actually existent 
objects. If simple numerical ratios were 
not to be found at least approximated in 
Nature, they could not be introduced into 




sculpture, because sculpture was showing 
forth the general type implicit in individual 
appearances. Just as in architecture many 
of the simpler ratios were abandoned for 
more complex ones which should be more 
appropriate, so in sculpture (but much more 
rapidly) artists appreciated the necessity 
of abandoning strict ratios of simple integ- 
ral numbers. But they seem still to have 
kept them as an unseen but operative 
frame-work of proportions. Nature and 
beauty strove for these ratios, but they 
were not attainable in earthly products; 
therefore the artist could not show them; 
but he must know them, and use them for 
his underlying structure, being careful to 
temper them in the interest of representa- 
tional truth. The temper at urae, of which 
Vitruvius speaks as imposed on architec- 
ture by the accidents of human vision, 
occur also in sculpture where they arc 
imposed by the accidents of material truth. 
This, I take it, is what Polykleitos was 
talking about in that other often-quoted 
and nearly always misinterpreted dictum 9 
'He said that the employment of a great 




many numbers would almost engender cor- 

rectness in sculpture" TO yap cv Trapa 

The sting is in Trapa /uKpoi/. By using 

an elaborate series of simple ratios it is 

possible to build up the coherent structure 

of an ideal figure; 10 but each one of these 

measurements must then be tempered a 

little lest the surfaces be not sufficiently 

true to natural forms. It is in these little 

shifts and changes that the secret of beauty 

lies. Without them, by a trifle you will 

fail of your object; you will have TO v 

Trapo, fUKpov. That is why we cannot 

recover the exact ratios when we take 

measurements of Polykleitan work; but 

we ought to get those ratios Trapa /u,iKpov. 

Obviously the fifth century sculptor did 

not hold that he was merely copying such 


and such an individual human being, con- 


light on the 

verting him into bronze or marble. He was 

true nature 

'constructing on the basis of natural appear- 

of Greek 

ances a series of surfaces and shapes and 

in sculpture 

lines which the eye could contemplate with 

satisfaction and the mind consider with ap- 

proval and which it would recognise as the 





appropriate perfection in fully developed 
human kind. The natural appearances 
of objects must, of course, be followed as 
a corrective for all extreme departures and 
as a starting-point for schematic or formal 
deductions; but there was no thorough- 
going insistence that concrete individual 
objects must be reproduced just as they 
looked and were. The statues of victors 
in the games did not and could not resemble 
those victors themselves. Later ages, 
struck by the absence of portrait-like qual- 
ities in these dedications, invented such 
idle explanations as the story, which Pliny 
gives us, that only an athlete who had won 
a victory for the third time could dedicate 
a statue with his actual form and features. 
Just as Pythios embodied his idea of the 
true form of an Ionic temple in the Athena 
temple at Priene, so Polykleitos embodied 
his idea of the true form of the nude male 
in his canonic Doryphoros. Pliny quotes 
Varro's complaint that the statues of Poly- 
kleitos were too much alike, paene ad unum 
exemplum; but that is just what we should 
expect of them, and the complaint is no 



more apposite than it would be to complain 
that all Doric temples are alike, 

To summarize: the "idealism," the 
"classic restraint," the "omission of non- 
essentials," so characteristic of fifth century 
Greek sculpture, are all traceable to the 
attempt to put into material guise an 
almost metaphysical abstraction, a type- 
form which should satisfy the reason in its 
quest of perfection and through the senses 
lead it on to attain the supersensual. 
This seemingly unsculptural and unreal 
trend originated in and developed out of 
the schematic forms of archaic art, which in 
turn grew out of psychological accidents 
wholly irrelevant to fifth century ideas. 
Here, as in all genetic processes, origin and 
validity must be kept distinct. To the 
Polyklcitan age the individual forms of 
line and surface were not memory-images, 
but the inherited alphabet of the sculptor's 
art, with which he was to spell TO KAAON 
or, better, 'O TIAI2 KAAO2. Although 
Nature was always the check and the 
corrective, the sculptor was not aiming at 
reproducing her chance individual appear- 




ances. He did not imitate TO, 
TO, Kara <J>\XTIV. His statues were not 
intended for replicas of unusually beautiful 
persons in unusually attractive attitudes. 
He did not make men as they appeared, 
but, in the deepest philosophical sense 
which his race could attach to the words, 
men as they were in their essence. 

This is precisely the judgement which the 
greatest of the later sculptors, Lysippos, 
passed upon his predecessors, if we are to 
trust Pliny's report "Vulgoque dicebat ab 
illis factos quotes essent homines a se quales- 
mderentur esse; u "He often said that they 
represented men as they really are, but he 
as they appeared to be." This sentence 
(like every other one touching ancient 
artistic criticism) has been consistently 
misunderstood. It has been imagined that 
Lysippos was distinguishing between real- 
ism and impressionism; but nothing could 
be more wide of the mark. We have only 
to look at fifth century sculpture to con- 
vince ourselves that, whatever the artists 
may have done, what they signally did not 
do was to represent men as they really are 



in Nature. Pliny's csscnt is Plato's TU> ovn 
ov and Aristotle's TO TI ty emu, which is 
not in the least like artistic realism or repre- 
sentational fidelity to natural appearances; 
and his mderentur refers to TO, <aivo/x,ei/a, 
which is the very thing which we nowadays 
call reality. There is no question of im- 
pressionism. "Ad -veritatem Lysippum ac 
Praxitelen access isse oplime adfirmant," 
says Quintilian; 12 and that ought to settle 

Let us consider for a moment what sculp- 
tural impressionism means. In painting, 
the incentive to impressionism arises out of 
the discrepancy between the hypothetical 
distance at which the painted objects are 
supposed to exist and the actual distance 
from our eyes at which they occur on the 
canvas, and out of the difficulty of suggest- 
ing luminosity correctly. If we merely 
paint on a reduced scale all the details 
which an object shows on close inspection, 
the modifications of perspective will not 
suffice to give us the illusion of seeing that 
object in its proper setting of atmospheric 
distance. W.e must paint, on the canvas 





close to our eyes, the look of that object at 
a distance, with all its blurred edges and 
qualities of colour and light. We must 
paint the retinal impression. This involves 
a complicated problem of colour values 
which will suggest these effects of light, 
since painting cannot reproduce them 
directly, but must evoke them by an 
equivalent construction out of her own de- 
vising. Aiming at greater fidelity to seen 
appearances, painting thus ceases to fol- 
low the materially "correct" outline and 
colouring of objects; and the opposition 
between realism and impressionism can 
thus arise. But in sculpture, where the 
true distances are accurately given and 
fidelity to colour and light is greatly subor- 
dinated to questions of surface-forms and 
outlines, impressionism in the sense just 
now indicated has no obvious place or 
meaning. There is, however, a secondary 
sense in which the term is sometimes used 
for sculpture. To show an object as it 
appears is in sculpture to show it as it 
actually is; but we might undertake to 
represent the object as it seems to someone 




who is looking hastily or casually or under 
emotional stress or under some bias of 
interest or judgement. Thus impressionism 
in sculpture comes to stand for something- 
very rnuch like emotional expressionism. 
It is not a normal "impression" such as the 
laws of optics and physiology would 
impose upon the normal seeing-eye; it is 
somebody else's impression, in the sense of 
a mental and emotional construction such 
as under proper stimulation we might be 
able to recreate in ourselves. Of all this, 
what is there in the work of Lysippos? 
A free use of massed shadow in the hair is 
the only impressionism that I can find ever 
ascribed to him; and the testimony of 
ancient writers and of extant works of his 
school and period point wholly the other 
way. When Lysippos said that he made 
men as they seemed to be, he was saying 
that he made them just as they actually 
looked. It was Lysippos who was the 

l( Ad veritatem Lysippum ac Praxitelen 

" Can this be true of Praxiteles? 

Critics have felt themselves at a loss be- 




(4) free 

;5) fine 
period, pre- 
ceding the 
ree period 

cause third century Alexandrian art, carry- 
ing on Athenian traditions, produced ultra- 
Praxitelean work and, at the same time t 
ultra-naturalistic genre. But there is no 
discrepancy or contradiction here. Instead 
of relying on the emotional power of ab- 
stract pure form, the Alexandrian school 
turned to direct pictorial presentation. In 
order to represent ugly, uncouth, bizarre, 
or ludicrous subjects, it copied these 
faithfully, down to every wrinkle and 
deformity: in order to represent beautiful 
subjects, it went equally far in the literal 
imitation of soft-textured skin, smooth and 
yielding surface forms, without edges and 
with veiled lines. It is simply a difference 
of subject-matter; the literal intent and the 
artistic method are in either case the same. 
For all his seeming idealisation of beauty, 
Praxiteles is the immediate precursor of the 
Alexandrian realists. 

When the schematic form disappears as 
the ideal norm of reference, intelligibly 
present and definable in every instance to 
which that form applies, when instead the 
artist copies the actual appearances of 




objects without deferring to a mental con- 
struction or generic prototype, art enters 
upon its highest career of technical achieve- 
ment. But its formal content is inevitably 
diminished and obscured; and at precisely 
this point in its career spiritual decadence 
sets in. In Greece this turning falls in the 
lifetime of Praxiteles (as in the sister art of 
dramatic literature it falls in the lifetime 
of Euripides). Lysippos is already of the 
decadence. For did he not admit that his 
predecessors imaged men as they were in 
their essence, but he himself merely as they 
appeared (very much as Sophocles was said 
to have shown men as they had to be; 
but Euripides, as they were). 

The reason why the decadence begins at 
the very moment of seemingly greatest 
achievement needs no elaboration for those 
who have followed our previous discussions. 
The true artistic aim of sculpture has prov- 
ed itself not to be this ignis fatuus of imita- 
tive dexterity and representational fidelity. 
To the sole, innermost, yet so frequently 
obscured, intention of the art as an art, we 
have already directed a detailed attention. 

This ends 
our inquiry 
into the 
form and 




If the argument for the true nature of the 

esthetic appeal and artistic purpose of 

sculpture be recalled, it will be obvious 

why, as estheticians, we must agree with 

the frequent opinion that the Hellenistic 

Age, however much it may mark technical 

advance, artistically is on the downward 

and our 
of Greek 

slope. "Cessavit deinde ars" says Pliny 
for the period after 296 B.C., and, except 


that we might be inclined to set the year a 

little earlier, we can hardly fail to agree with 

his pronouncement. 








I MUST presuppose that the history and 


achievements of Greek architecture are 


sufficiently familiar to the reader. He will 

know, for example, that the dominant form 

or theme is the colonnaded temple; that 

this is built with wall, column, and 

horizontal beam; and that the structural 

elements are the square-hewn, close-fitting, 

mortarless ashlar block, and the various 

members of two distinct and characteristic 

"Orders," the Doric and the Ionic, to the 

latter of which is allied a sub-Order, the 

Corinthian, scarcely differing from the Ionic 

save in the form of its capital. He will 

know, further, that there are temple remains 

dating from the sixth century B.C. and 

remarkable for their vigour and simplic- 

ity; that in the fifth century were achieved 

the refined and powerful buildings on the 

Athenian acropolis, in which the Doric 





and popular 

Order reached its formal perfection and the 
Ionic its first flowering on Attic soil; and 
that in the two succeeding centuries the 
Ionic style was perfected by Pythios and 
Hermogenes and their respective schools, 
on the west shores of Asia Minor in such 
notable exemplars as the Mausoleum, the 
Athena temple at Priene, the Artemis 
temples at Ephesos and at Magnesia on 
the Meander, and the long-travailled but 
never-completed giant temple of Didymean 

I must also assume that most of the tech- 
nical vocabulary of the architect will con- 
vey its legitimate meaning to the reader, 
and that he has already noticed the obvious 
and salient virtues or limitation? of the 
Greek style. 

Thus he will know that it is a current pre- 
cept that Greek buildings always express 
their purpose and the mechanics of their 
construction. A temple-plan is self-explan- 
atory; its elevation never misleads us with 
any pretence or disguise. The entablature 
is not carved or moulded to produce a 
merely decorative facade: the epistyle has 



horizontal stressings to show its function; 
we may guess that ceiling beams bed behind 
the frieze; the cornice is unashamedly a 
water-shed; the pediment occurs of neces- 
sity because a sloping roof can end in no 
simpler or more obvious manner. The 
akroteria are the only structural non- 
essentials and I am not altogether sure 
that these may not have been survivals of 
weights to hold down the roof, like the 
stones on Swiss chalets, or the interlacing 
of rafters. 

We can scarcely acquit the Greek archi- 
tects of having followed a third popular 
modern precept by making their architec- 
ture express its material, since they not 
merely painted their marble, but in the 
members of the two Orders carved forms 
(such as triglyphs, dentils, guttae, regulae) 
which were clearly survivals of some tim- 
ber construction. Yet they exploited the 
good qualities of their building material, 
by giving their marble walls that lustrous 
finish which still endures upon them, and 
by carving designs whose sharpness of line 
and shadow-free shallows emphasised those 






beauties which reside more in fine marble 
than in any other stone. On the other 
hand, they could not tolerate rustication or 
ruggedness (except for fortification walls 
and similar uses) but covered their coarser 
materials with a stucco of powdered marble 
which effectively gave the lie to its native 
qualities. That they sheathed exposed 
timber with terracotta or bronze cannot be 
instanced against them, since it was done 
not so much to disguise the material as to 
protect and preserve the wood. 

There is a fourth precept to which good 
architecture is generally supposed to con- 
form. It must express (it is said) the men- 
tality and civilisation of its builders as 
though whatever man consciously makes 
could fail to be eloquent of himself! Even 
a Moorish tower upon a Renaissance palace 
built for the servants' hall of an American 
millionaire's country-place is eloquent of 
its age and culture. Eclecticism is as much 
a style as any other and as significant of 
its attendant civilisation. The sky-line 
of Fifth Avenue is very eloquent of the per- 
turbed and composite culture that made it 




and enjoys it. But it is intended to assert 
by this particular precept, that the spirit- 
ual qualities of an age should somehow be 
deducible from- the buildings which it pro- 
duces; and it is clearly true of Greek archi- 
tecture that it is ordered and logical, that 
it harmonises part with part, and parts 
with whole, and that, to a most remarkable 
degree, it is unfantastic and, in all its ele- 
ments and proportions, always implies a 
human use or service, and a mortal eye 
and a rational mind to observe and approve 

There are no structural irrelevancies in 
Greek building, such as columns which are 
not the true supports of their superstruc- 
tures, or fagades which merely mask the 
disparate structure behind them. But 
though there are no structural irrelevan- 
cies, neither is there much structural inven- 
tiveness or ambition. The Greek temple is, 
in its intent, a one-story affair. The super- 
posed colonnades in its interior were forced 
on the builder by the physical necessity of 
supporting the roofbeams and the esthetic 
impossibility of doing so by means of a 

of Greek 




single story of columns (since these would 
have to be taller and so more massive than 
those of the exterior order). The two- 
>tory colonnade with ground-floor and gal- 
lery, thus produced, was often employed 
to line market-places and open squares, 
but seems not to have incited builders 
to attempt a third or a fourth stage of 
columns. The tower-like superposition 
of the orders was reserved for Roman 
facades and the restless Renaissance to 
achieve. The stoa, thus limited by cus- 
tom to two stages, was further restrained 
in its height by the characteristically 
Greek consideration that a column more 
than 12 feet high would seem out of pro- 
portion to the human beings for whose 
service and convenience it was intended. 
Here, as so often in things Greek, it seemed 
TTO.VT avBpa Trdvrwv xprjfJLOLTwv fttrpoy uvai. 
The temples, being houses of gods or 
at least of their images, were not cribbed 
and cabined by these demands of human 
proportion: their single story of columns 
(you would imagine) could swell to any 
height. But here another proportion 



stepped in to check their rise toward 
the colossal: the distance from column to 
column always bore some sort of estab- 
lished relation to the height of the shafts, 
and columns could not be too widely 
spaced lest they could no longer be safely 
bridged with horizontal beams of stone. 
The Greeks were timid engineers: 1 a 15- 
foot air-span gave them pause, and the 
lintel of Diana's temple at Ephesos was a 
gigantic achievement. And so when they 
strove toward the colossal in height, their 
columns were allowed but barely sixty feet. 2 
Except for such exceptional forms as the 
lighthouse of Alexandria and the Mauso- 
leum of Halicarnassos, how many Greek 
buildings (like the Zeus temple at Girgenti) 
overpassed a height of one hundred feet? 
The theatres, to be sure, stretched up till 
their topmost rows might be more than this 
distance above the dancers' circle; but the 
seats were all bedded safely on the sloping 
hillside. For the Greeks before the Roman 
era, no hillside no theatre. Nothing is 
more surprising than to sketch the eleva- 
tion of one of the great French Gothic 




cathedrals and to the same scale superpose 
upon it the facade of any of the greater 
Greek temples. Where the two buildings 
chance to agree in width, as do the Parthe- 
non and the cathedral of Chartres, it is all 
the more astonishing to see the Gothic 
pushing its nave to twice the Greek 
temple's height before ever it begins to 
stretch its towers and spires toward the 

Nor were plans varied or complex or 
ambitious. 3 The temple was the god's 
house and the depository for his posses- 
sions. What then should a god do with 
more than two rooms? When the cult- 
statue was housed and the dedications and 
sacred treasures were safely sheltered, noth- 
ing remained to do. A surrounding colon- 
nade to accommodate the god's worship- 
pers, perhaps; but nothing more. The 
altar, if only because of its smoke and fire, 
was in the open air and escaped housing. 
A peribolos wall, marking out the god's 
acreage, often shut in indiscriminately 
temple, altar, and ex voto dedications. The 
Greek architect had no idea of the myster- 



ious in setting or approach : a few chthon- 
ian shrines in caverns is all that can be 
cited. Nor did it occur to him that his 
god might house, dark and inaccessible, 
beyond great walls and many rooms, 
approached through courtyard and hall- 
way and stair, as the gods of the Pharaohs 
lived in the penetralia of Karnak or Luqsor. 
Nor were there kings to dwell in magnifi- 
cent palaces. Even a sixth century tyrant 
was a man among men. Peisitratos might 
take the Acropolis for his house, and Diony- 
sios might have his dwelling on Ortygia; 
but these were fortresses, not palaces. 
Even the princes of Hellenistic Pergamon 
have only a paltry bit of the great hillslope 
for their private places. The "palace" of 
Pericles? There was none. And the great 
fifth century Athens, which was responsi- 
ble for the masterpieces of the Acropolis, 
was only a TroAts KUKWS e/opv/xoTo/x^/xeVr; 
of small mud-brick houses and crooked 
narrow streets. The greatness of Greek 
architecture is certainly not domestic, but 
lies in temples and public buildings of the 
polls. Yet how monotonous are the plans 




of the Greek temples! and although their 
settings are often wonderful on the windy 
sunrise cape of Sunium, among the waste 
grey folds of Arcadian Bassae, under the 
shining frown of the Delphic cliffs how 
much reliance is placed on Nature and how 
little does Art contribute! Delphi is as 
idly and irrelevantly marshalled as the 
accretions of a cemetery; Delos is an archi- 
tectural pepperpot; Olympia is unconsid- 
ered and fortuitous; the Athenian acro- 
polis, a crown of splendour upon a head 
of tangled hair owes much of its 
effectiveness to the lofty, clear-shining 
aloofness of its site. Neither in the 
planning of the individual building nor in 
its relation to other buildings does the 
Greek architect show invention or artis- 
tic imagination. I do not feel confident 
that he even considered this latter 
relation unlil the fourth century: the 
symmetry a pen pres in the arrangement of 
the buildings on the Acropolis ma^'be very 
subtle or very haphazard. 

Mnesicles' unfulfilled design for the Pro- 
pylaea, the town-plan of Halicarnassos and 



perhaps Rhodes, the carven giants against 
the gigantic masses of the Zeus temple of 
Akragas, the Erechtheum plan with its 
imperfect reaching-out toward a freer 
form 4 a list of six or seven such items seems 
to exhaust the instances of original and 
powerful departures from the lucid but 
easy perfection of an established tradition. 
Two "Orders" and a variant only that to 
show for the activity of centuries! Two 
entablatures and three kinds of capital is 
that fecundity? is that inspiration? Is it 
then so hard for the human brain to invent, 
so impossible for it to create, that one of the 
most intelligent and esthetically most sen- 
sitive of races should have only this to 
show? They are very fine of their kind, 
these Orders and these temple-, colonnade-, 
tholos-, and theatre-forms; but under what 
limitations and abnegation of originality 
was their perfection wrought! And how 
could so many generations of active archi- 
tects remain content to copy and re-copy 
from the same meagre repertory, as though 
there were no other way to build, no other 
plans or forms, nor even any other mould- 


An archi- 
tecture of 
fixed and 






ings to be imagined, and 
a proportion or to refine 
line were achievement 

as though to alter 
a subsidiary out- 
enough for any 

How can we explain such absence of 
individual invention and artistic variety? 
Only by referring it to a deep-seated prej- 
udice of the Greek mind. To explain this, 
there will be need of a short digression. 






Under the perspective of time, the dis- 
putes of philosophic schools are apt to 
take on an appearance of fundamental 
irrelevancy. The real issues are usually 
taken for granted because they house so 
deep in the general prejudices and presup- 
positions of the times. Aristotle pulled the 
Platonic Theory of Ideas to pieces, only to 
assert a very similar doctrine of forms in 
its place. From this it is fair to conclude 
that the assumption of visual prototypes 
for the various kinds or species of existent 
objects to believe, for example, that the 
abstract concept "mouse" implies an actual 
mouse-form universally inherent in all 
mice and controlling their shape and ap- 
pearance came naturally and as it were 
necessarily to the Greek mind. Modern 
scholars tend to assume that the Greek 
philosophers derived these "Forms" out 
of the logical concept and out of the lin- 
guistic necessity of using a single word to 
name all the objects of a given class (as 
when we call all mice by the same word 

The Theory 
of "Ideas" 
or Canonic 




mouse"}. They look on them as a direct 
outcome of the behaviour of human thought 
and the analysis of human speech. But in 
addition to all this, the Greek philosophy 
implies a trick of visualising, of appre- 
hending the characteristic shape and line 
by virtue of which all mice, for example, 
look sufficiently alike to be recognisable 
instances of their species. For the Greek 
mind there was always a concrete visual 
image somehow attached to the abstract 
universal concept. We ourselves use such 
words as "concept," "abstract term," and 
similar expressions which do not have any 
very clear metaphorical content; but Plato 
talked of iSu (things tSeti/ to see) or of eiSvy, 
which again are things seen; even the 
ordinary Greek word for "knowing" (o!8a) 
had in it this same root of "looking" or 
"seeing." And most of the logical pitfalls 
by which the Platonic Theory of Ideas was 
beset are due to the concrete hard-and-fast 
visible quality (and therefore existence) of 
which the 187; could not divest themselves. 
Reference has often been made to the 
peculiar qualities of light and atmosphere 



in Greek lands, to that clearness which 
robs distance of its atmospheric mantle and 
that luminosity which deprives shadows of 
their power of plastic illusion, so that the 
eye sees in terms of contours bounding flat 
surfaces, and the brain remembers in terms 
of simple visual forms, of areas with sharp 
outlines and consequently very definite 
shapes. If we remember, too, that the 
Greek intellectuality (partly because of 
this extreme atmospheric lucidity under 
which the world of sense was presented to 
it?) was not "cloudy-minded" and "un- 
earthly," but markedly observant of the 
external world and analytic of its phenom- 
ena, it will seem natural that such a race, 
highly intellectualised in the direction of 
observation and analysis, and immersed in 
a sense-world of highly accentuated out- 
lines but rather diminished tactile and 
plastic values, should imagine its concepts 
as visual linear forms. Not the mass, not 
the material constitution, not the physical 
behaviour, but the seen appearance, was for 
them the essential. To know what a thing 
is, they must know the look of it. The 



its in - 



material might be comparatively irrelevant, 
since objects can be of different substance 
and yet have the same look; and matter 
changes so illusively (as when food turns 
into flesh and blood), while the form is 
abiding and amenable to the understanding. 
The matter of a man may suffer all manner 
of metabolism, but the form undergoes no 
such change; if annihilated in one indi- 
vidual by death or altered by mutilation, 
it reappears unchanged and undiminished 
in others of the species. Surely, if the 
Greek mind had not, instinctively as well 
as consciously, thus regarded the world of 
objects as a series of typical forms dis- 
played and embodied in individual in- 
stances, Plato could never have imagined 
his theory of visual-forms or found an 
audience to which to expound it. 

The objects which man makes his im- 
plements, his places of shelter, his fur- 
niture these too come ranged in species, 
and each species has its characteristic 
form whose specific shape and appearance 
depend largely on the purpose or use of the 
object in question. The shape of a mirror 



is not arbitrary: its purpose implies that 
it should be the shape of the human face, 
its use suggests that it should have a 
handle by which it can be grasped and 
which will also serve it as a stand, for 
which purpose the handle should have a 
stable base. Thus defined, the necessary 
shape and appearance of a mirror lie within 
rather narrow limits; but among the vari- 
ous shapes which would satisfy the fore- 
going requirements, are not some more 
necessary" than others? Is not the 
circle the necessary prototype of all face- 
shapes? For the handle and stand, will not 
certain relationships in the measurements 
and outline recommend themselves by the 
simplicity or coherence of their mathe- 
matical and logical properties, so that they 
will seem more nearly "right"? And do 
we not end with a conviction that even to 
such a man-devised thing as a hand- 
mirror there is a single fitting and perfect 
form which man can discover by experi- 
ment and thought? There is something of 
the eternal and super-mortal about a form 
which is thus attained; it seems as though 





(like the truths of mathematics or logic) it 
must all the while have been in existence, 
implicit but unidentified. Almost it would 
appear as though the artist himself invents 
nothing: he merely sees more clearly than 
other men how things should be. The 
more he approximates perfect rightness in 
the forms of the things which he makes, the 
more nearly he has learned to materialise 
what the Divine Necessity has already 
constructed. . . . "The craftsman, look- 
ing toward the true form, so fashions" 
(6 8tjfJLLOVpyo<s Trpos T^V I8eav /3A.7rwi/ ovrw 
Troiei 5 ) said Plato of such man-contrived 
things as couches and tables. 

How then does the artist's activity differ 
from that of the philosopher, except that 
one gives material while the other gives 
verbal expression to the concepts which he 
discovers? It is the good craftsman's 
business to seek out the right form, which 
is the true and perfect species-type for the 
kind of object which he is making. (And 
perhaps beauty and intellectual fitness will 
attend his discovery, because the true and 
primal types are god-made.) 




In Greek pottery, species appear and 
maintain themselves with the most start- 
ling defmiteness of shape. Lekythos, 
pyxis, aryballos, hydria, krater, psykter, 
skyphos, amphora, kylix, kantharos how 
distinct and distinctive they all are, each 
form rigorously determined by a consid- 
eration of its purpose and usefulness, with 
its own secret of grace of line and harmony 
of part! One comes to believe in their 
real objective existence as firmly as though 
they were genera and species of the animal 
world evolved by Nature. One can detect 
this same spirit in all the creations of the 
Greek genius: clothes, household furni- 
ture, tableware, all fall into type-forms 
characterised by an almost logical co- 
herence and exactness. 

The work of the architect will reveal a 
similar trait. If he plans a temple, his 
most conspicuous artistic duty will be to 
consider afresh the adequacy and Tightness 
of the temple-form as his predecessors have 
sought it out and embodied it. What 
criteria can he use for scrutinising their 
success or failure? Only those of that 

in vase- 

in archi- 




nner necessity by which a form is recog- 
nised as right and true. Whatever is un- 
ssential, whatever is not necessary to the 
temple for its purpose or stability or dura- 
bility, is irrelevant to the form; it is not a 
part of the form, it is not a part of the 
temple. The expense and labor and time 
incident to the processes of building con- 
spire to encourage the architect in elim- 
nating the superfluous. How then could 
there logically be different temple-plans, 
when there is only one purpose for which 
temples are used, one identical force of 
gravitation to combat and master, one rain 
and sun to oppose? There should only be 
variety insomuch as there might be inade- 
quacy in expressing, or incompleteness in 
discovering, the true and right form. The 
Parthenon is not a different building from 
the temple of Poseidon at Paestum: it is 
the same building grown closer and truer 
to its proper and perfect semblance. 
Iktinos invented nothing: he merely saw 
more clearly how things should be. 

But surely individual taste must enter 
in? the architect must arrange and vary 




details to suit himself or the preference of 
his patron? That might be the case, did 
perfection depend on artist or public; but 
neither of these can create right relations, 
they can only seek to embody them. One 
can tune the strings of a lyre to yield any 
pitch whatever; but the concords and dis- 
sonances do not depend on the musician or 
his public, but on the mathematics of the 
musical scale. Individual preference is but 
a chance guide to help the sensitive crafts- 
man to the apprehension of those necessary 
and inherent harmonies which mark out 
the true form from all more or less imper- 
fect approximations. 

The rivalry between Pythios and Her- 
mogenes, who lived about a century apart, 
was a contest to discover the same thing 
the most just proportions for a series of 
architectural parts of established shape in 
predetermined sequence of arrangement, in 
fact the perfect norm for the century-old 
Ionic Order as applied to the century-old 
plan of the peripteral temple. And these 
were great architects, the greatest perhaps 
among the Greeks of their age, not because 




they had done something new, but because 
they had done something very old a little 
better than anyone else had ever managed 
to do it. The Athena temple at Priene and 
the Artemis temple at Magnesia were 
claimants for one and the same title, that 
of the Ionic temple par excellence; their 
size, their use, their location, their cost 
were all minor considerations to this great 
distinction of being canonic, the perfection 
of their tribe and kind. In comparison 
with the differences which are apparent in 
any two Gothic cathedrals built within the 
same century A. D., these two Ionic 
temples would be indistinguishable one 
from the other. In this Ionic architecture 
of the Asia Minor coast we can obtain a 
very admirable notion of the extent to 
which individual preference and invention 
were encouraged or allowed to assert 

The impression that the Greek architects 
were merely seeking the right form, not 
striving to be original or emotionally ex- 
pressive, occurs insistently. The propor- 
tions of the Doric Order change rapidly 



but consistently during the sixth and fifth 
centuries B.C. With the attainment of the 
right form in the Athenian Propylaea the 
process of development is almost entirely 
arrested. Later, the decidence of the 
Order from favour for temple-construction 
and the obvious need of a lightened en- 
tablature in the use of this Order for 
colonnades, entirely altered the ratios: the 
form was no longer right, an altered use 
had modified it. The profile of the Doric 
capital changes greatly during the sixth and 
fifth centuries; but after the time of the 
Propylaea the only important change is 
the slight straightening of the top of the 
echinus which makes the capital so much 
easier to cut and entirely robs it of beauty. 
When the architects no longer appreciate 
Tightness of form, the Practical and the 
Labor-saving are allowed to announce their 

Until Hellenistic times there was only 
one important innovation among archi- 
tectural forms the Corinthian capital 
and that was produced under logical pres- 
sure because of the inadequacy of the exist- 




ing forms. The Ionic capital posed an in- 
soluble problem at the point where a 
colonnade changed direction (as at the 
corner of a temple), while the Doric cap- 
ital could not be extricated from its use with 
the Doric entablature whose triglyph was 
a tyrannous nuisance with its interference 
in the free spacing of columns. Moreover 
neither the Doric nor the Ionic capital 
strictly satisfied the logical requirements 
for the right form. The capital's function 
is to take the entablature weight and trans- 
mit it to the column-shaft. As an inter- 
mediary its shape should be consonant 
with those which it serves: it should be 
rectangular above and circular below in its 
cross-section, and in the intervening part 
it must perform without effort the transi- 
tion from rectangle to circle. The Ionic 
capital does not meet these requirements; 
the Doric lacks the transition; but the 
Corinthian will withstand the most rigor- 
ous criticism. It is a marvellous meta- 
morphosis of a circle into a rectangle by a 
conventional device of leaves tied around 
a shaft. 



The development of the Ionic column- 
base from the crude and unfelt forms of the 
Samian Heraion to the Attic or the fourth- 
century Asia Minor types is a coherent and 
gradual process. Contemplating it. we 
have the impression of beholding in opera- 
tion that which the metaphysicians call a 
Final Cause, as though it were not each 
term in the series which occasioned the 
next, but the ultimate and perfect form 
like a magnet exerting its attraction 
through all the attempts of all the earlier 
builders. Compare, in particular, the 
Ionic bases of the Propylaea, of the east 
porch of the Erechtheum, and the north 
porch of the same building: the last form 
lies in the first, and progress is as un- 
hesitant as though the path were known 
and the goal sighted in advance. Equally 
impressive is the abrupt cessation of ex- 
periments and changes as soon as the 
right form has been attained. The pro- 
file of the north porch bases recurs some 
eighty years later on the monument of 
Lysicrates, almost unmodified, and the 
form seems to have imposed itself upon 




the ancient builders almost as authorita- 
tively as upon their modern imitators. 
The Corinthian capital has much the same 
history; its form, once reached, undergoes 
little further change until late Hellenistic 

This explanation for the extraordinary 
tyranny of the canonic form is manifestly 
inapplicable to all the elements of the 
Orders. The shape of a triglyph or mutule 
could not be deduced from the necessities 
of a roofed two-room shrine in trabeated 
stone or from any other intellectual for- 
mula or requirement. One could derive 
the column logically, but not its diminu- 
tion; the capital, but not its profile; the 
architrave and perhaps the frieze, but not 
their scheme of decoration; the cornice, 
but not its profile. With the substitution 
of stone for timber in very early times (for 
however much one may differ as to the 
precise structure of this timber prototype, 
it is impossible to refuse credence to it in 
some form or other) the structural reasont; 
for the appearance and pattern of regula, 
guttae, triglyph, mutule, fasciae, and 



dentils disappeared; but the things them- 
selves remained. By immemorial tradition 
they survived, and just because strict logic 
made no demand for their existence, it 
made no demands about them at all. They 
had come into existence and now persisted 
in their own right as established species 
much as there might be no discoverable in- 
herent necessity for owls or weasels, but 
there they were, and each had its distinc- 
tive and established form. That is why 
throughout the classic period these ele- 
ments of the Orders found no rivals. The 
mind of man might create all manner of 
designs and decorative details; but these 
would have no inner necessity, no reason 
for being thus rather than otherwise. 
Dentils and mutules and triglyphs might 
be equally arbitrary; but there they were. 
They existed. As detail of some sort was 
clearly needed to enliven and lighten the 
superstructure, these decorative forms in- 
herited from the old timber construction 
were retained. So it was that Greek archi- 
tecture had only two Orders. 




br the 
;enacity of 
tural forms 

Such an explanation, however, is not 
wholly satisfactory. In other arts, decora- 
tive ornaments usually have a very well 
marked career and run through a history 
with recognisably similar stages. The 
ornament begins as a more or less skillful 
and accurate imitation of some actually 
existing object of the animal or vegetable 
or even inanimate world. By dint of 
repetition it loses its fidelity and freshness, 
becomes stereotyped and conventionalised, 
and after a time may drift so far from its 
original appearance that its representa- 
tional intent may be lost to the craftsman, 
who blindly reproduces a meaningless pat- 
tern. (The flower and marine motives of 
Cretan, Mycenean, and sub-Mycenean art 
may show the phases of such a develop- 
ment.) But in Greek architecture (though 
we find such naturalistic and living imi- 
tations as those of the Corinthian capital 
or some of the rinceaux of the simas, as 
well as such ultra-conventionalised orna- 
ments as those of the leaf -patterns on the 
various cymatia), both naturalistic and 
conventionalised ornament, once they have 



been accepted for architectural use, undergo 
remarkably little further change. A prac- 
tised eye can distinguish between a Lesbian 
cymation carved in the sixth and one 
carved in the third century B. C.; but as 
far as representational value or general 
pattern is concerned, the difference is 
slight. Only a specialist can tell in what 
century a given triglyph was carved or de- 
tect the difference between Ionic column- 
bases of the fourth century B. C. and those 
of the subsequent centuries. Why did the 
ancient architects cling so closely to forms 
and patterns that were neither necessary 
nor especially logical, but merely tradi- 
tional and accepted fashions of the school? 
Perhaps it would be pertinent to ask a 
closely similar question of the architects of 
our own day. Why are the old Greek 
Orders still retained to-day and why are no 
new Orders invented to supplement or sup- 
plant such hackneyed forms and formulae? 
When so much might be created, why only 
an uninventive repetition of a fashion that 
every eye must have seen a thousand 




times? In a word, why does architecture 
persist in having " Orders" at all? 

Why, after so many hundred or even 
thousand years, has architecture so few de- 
vices in her copy-book? Is it really so 
hard to invent an Order, that we have only 
five or six for common service? And what, 
fundamentally, is this fetish, this jargon of 
the initiate, about "styles" a Gothic 
" style," a Romanesque, a Greco-Roman, a 
pure Hellenic "style"? Who or what 
commanded that architects must keep to 
style, must copy and imitate and adapt, 
but not invent wholly new things for them- 
selves? In literature plagiarism is held to 
be anything but a virtue and a pastiche is 
accounted worse than a "potboiler." Yet 
a wholly original, inventive, and unplagiar- 
ising architect would never get a commis- 
sion. How came such a deplorable con- 

It is perfectly true that examination of 
the historical process in which architec- 
tural styles and orders have come into 
being should warn us that the attainment of 
perfection is a tedious thing requiring the 



combined and successive energies of many 
men and outlasting the span of a single 
human life, so that it is the part of caution 
and sound sense to take from the past 
what the past has found good; but I think 
there is no other great art in which such a 
plea can be advanced or entertained. No 
doubt it is hard to write a good new piece 
of music; but the modern composer can- 
not therefore plead for privilege to take all 
his themes from Haydn and Mozart and 
his melodic developments from similar 
treatments by earlier masters. 

Of course it is immensely convenient to 
trace off a good bit of detail from the Arch 
of Titus or the temple of Castor and Pollux; 
or to put the doorway of the Erechtheum 
in toto for the main portal of a modern 
bank; but if convenience is the only excuse 
for such a practice, it is high time for a new 
generation of practitioners to arise. Clearly 
there is some fundamental principle which 
is operative in architecture, however out 
of place it may be in the other arts. 

Contemporary critics often bewail what 
they term the impasse of eclecticism and 





ture a 
tional art 

pretend to see no hope for an art which 
makes past achievement its sole guide, 
ounsellor, and friend. They have failed 
to observe that the tenacity of architec- 
tural tradition is due to an instinctive 
clutching for something permanent, fa- 
miliar, and universal, on which to base 
diversity and artistic interest. The paucity 
of invention, the seemingly suicidal re- 
straint of variety, are necessary because 
architecture is seeking to establish for itself 
a real world of recognised and recognisable 
objects. Painting has the seen world to 
draw upon: it does not have to create and 
establish its trees, rocks, streams, mead- 
ows, and animals, or waste any effort in 
persuading us that they are things which 
we already know. But architecture has to 
invent its world of objects first, before it 
can use them. And so, by the preference 
of generations of men, it settles upon those 
types and patterns which most commend 
themselves for their grace or expressive- 
ness or utility or simplicity, and these be- 
come its world of real objects which it 
imitates and represents amid those formal 




relations which physics will permit and 
good taste commend. The fundamental 
in such a theory is the assertion that, in 
spite of appearances to the contrary, archi- 
tecture tends to become a representational 
art, and that it imitates or represents a 
conventional world of its own creation. 
It is precisely because a Doric entablature 
is already familiar that it is artistically 
useful. The architect does not invent new 
orders for much the same reason that a 
writer does not invent new words, but uses 
the ones which are already known to his 
public, or that the painter does not invent 
botanical, zoological, and geological worlds 
of his own, but imitates the familiar real- 
ities of the ordinary world of sense. It 
should follow, as a logical consequent, that 
the artistic intention of the architect will 
be unintelligible to anyone who is not aU 
ready conversant with the traditional and 
conventional forms which he reproduces; 
and this is actually the case in direct pro- 
portion to the fixity with which the con- 
ventional forms are established. To the 
uneducated public the language of the 




and there- 
fore in need 
of conven- 
tional form 
for its rep- 

classical orders is wholly lost, precisely be- 
cause they are unaware of its alphabet and 

Greek architecture, indeed, is the out- 
standing instance of this practice of estab- 
lishing an artificial language by convention, 
in order to communicate architectural emo- 
tion. Gothic seems here (as in so much 
else) to have been of the other school and 
persuasion; -yet for all its inventive free- 
dom in profiles and shadows for the mould- 
ings of piers and doorways, and its fertility 
of intricate decoration, Gothic seems to 
have been well aware that its freedom lay 
within very definite bounds and to have 
been scrupulously careful of the appropri- 
ateness of form to function. Gothic was 

rich, picturesque language: the Greek 
Orders were as concise as the Greek philo- 
ophical vocabulary and consequently 
equally capable of subtle distinctions. 
Only he who knows the Greek Orders very 
intimately can really appreciate a master- 
piece of Greek architecture. Though in- 
;ellectually we may understand the in- 
tricate minutiae which occupied the build- 




ers' attention, perhaps very few to-day can 

really feel the emotions which these min- 

utiae are intended to impart. They are 

like the subtleties of language which make 

the magic of great poetry: we must know 

the language amazingly well or we shall 

wholly miss this magic. 

* * * 

In discussing the ideal trend of archi- 


tectural form, I have argued as though the 

is there 

eidos or right form were a changeless 

for these 

thing. But this is of course no more true 


of man-made forms than of the forms of 


the natural world with their constant 

though inappreciably gradual evolution. 

Just as in the realm of plants and animals 

the species-type changes, so in architecture 

(but much more rapidly) the various stand- 

ards change and evolve under the influence 

of taste. So definite and so rational is 

this transformation of the Greek Orders, 

that it is almost possible to date any given 

building merely by assigning its place in 

this evolutionary process. 6 

But if the form is dependent upon con- 





temporary taste and subject to constant, 

if gradual, change, what assurance could an 

ancient architect have that he had indeed 

found the right and true form? Since all 

the elements of his work were ultimately 

man-devised and man-perfected, what pos- 

sible guarantee of their objective fitness 

could there be, or what sanction for their 

claims to be the best type? That is a ques- 

tion which the Greek must have asked him- 

self. He found the answer just where we 

to-day might wish to find it in science, 

though of course it was the science of his 

day and generation and consisted mainly 

of geometric theory. 





early in their career in the course of 
Pythagorean and other geometric inves- 

It is an impressive discovery when the Anintci- 
human mind first catches glimpse of the lectual 


eternal supersensuous laws ruling the f rom tn e 
seemingly casual appearances of the world super- 
of sense. This moment came to the Greeks sens " al . 

world, oi 
(Cf. fifth- 
.. .. T 1 0.1 -A century 

tigations. In musical theory its appear- sculpture 
ance was most striking. Sounds those ch. in.) 
intangible and invisible occurrences, seem- 
ingly unruled by anything but a fortuitous 
concordance among themselves suddenly 
admitted their allegiance to the tyranny 
of geometry and number. The consonance 
of two notes was shown to depend on the 
presence of a simple integral numerical 
ratio between the lengths of the strings 
which produced them. Beauty and ugli- 
ness of sound were but functions of Num- 
ber and ratio, and therefore founded on 
something measurable, something intel- 
ligible. Everywhere, order showed its con- 
trol within the universe in the paths of 
the stars, in the structure of material 




things and everywhere, order seemed to 
be traceable to the influence of Number. 
If the right and wrong in something seem- 
ingly so elusive and unmeasurable as mu- 
sical tone was based on Number, was it 
not even more probable that a similar geo- 
metric or arithmetic basis should deter- 
mine right and wrong in the appearance of 
seen objects, which were material things, 
measurable and directly amenable to geo- 
metric notions? .Nature is orderly. The 
forms for which she strives are strikingly 
symmetrical and numerically rational. 
The accidents of matter obscure arid con- 
fuse the simple geometry of her intentions; 
but if we compare enough specimens of any 
species, we can eliminate the individual ac- 
cidents and construct the true form. Here 
then is a cardinal assumption of Greek 
esthetic practice that there is a true form 
for every class of objects and that such a 
true form is characterised by its geometric 
simplicity, by the commensurability of its 
component members. For if its parts be 
not simply commensurable, then complex 
and therefore less perfect numbers will 




enter and take the place of the more per- 
fect ones which might have been employed. 
Commensurability of parts, <rv//,/xeT/>ia, is 
consequently a test of Tightness of form, 
and it behooves the architect to work by 
the aid of its precepts. 

The generative ratio of the Doric Order 
was fixed as 2 to 1 in the earliest times of 
which we have knowledge. In the round 
building whose architrave was discovered 
in the foundations of the Sikyonian treas- 
ury at Delphi this ratio was apparently not 
followed: in Hellenistic times the entab- 
lature (because so diminished in height) 
is frequently poly-triglyphal. But be- 
tween these limits of time, the ratio of 2 to 
1 served for what may be termed the gen- 
erative formula of the Order in its hori- 
zontal extension. To every normal column 
span we find (enumerating the elements as 
they lie one above the other) 

1 column, 

2 regulae, 

2 triglyphs + 2 metopes, 

4 mutules, 

4 lions'-heads, 

(8 rows of cover-tiles?) 




so that an ordinary Doric colonnade beats 
out a rhythm of full-notes in the columns, 
half and quarter notes in the entablature, 
and (frequently) eighth notes on the sloping 
roof, and even sixteenth notes in the floral 
ornamentation of the sima between the 
lion's-head water-spouts. In the Zeus 
temple at Olympia the measurements in 
ancient feet show how thoroughly and co- 
herently the ratio was applied: the width 
of the roof-tile is taken as two feet, twice 
this measure gives us the width of a mutule 
with its via as well as the distance from 
lion's-head to lion's-head on the roof gutter, 
twice this latter measure gives the width 
of a triglyph with its metope, as well as 
the length of an abacus block, twice this 
measure gives the width from column 
center to column center, which is also the 
length of an epistyle block, and twice this 
last measure gives the height of a column. 
Finally, if the central akroterion of Nike 
was life-size, the total height from stylo- 
bate to peak of gable-figure may very prob- 
ably have equalled twice this last measure- 
ment. And so with 2X2X2X2X2 



(X 2?) the exterior of the temple is built 
up, and the tacit assumption seems to be 
that the eye, contemplating these forms, 
will be rhythmically affected by the simple 
.... numerical ratios inherent in them. 

The giant columns of Persepolis clearly 
betray Greek influence, and not least in 
their careful observation of numerical sym- 
metry. The two fore-parts of bulls which 
carry the rafters are supported by a four- 
sided theme, each side of which has four 
scrolls or volutes (each decorated with 
sixteen-petalled rosettes) into which a four- 
filleted band has been wound; this in turn 
is supported by a capital whose upper por- 
tion shows eight, while its lower portion 
shows sixteen, floral divisions; below, a 
shaft with forty-eight flutings is supported 
on a base whose petal forms number 

There can of course be no doubt of the 
conscious use of this simplest of all nu- 
merical ratios as a generative formula in 
the Doric Order, but it should not be im- 
agined that all the measurements therefore 
will be found to be mathematically exact 




all equal or double or quadruple in the 
ancient Doric temples. In the Sicilian 
temples' especially there occur the greatest 
irregularities in the spacing of the columns 
and the sizes of the various members. 
Much of this may be attributed to indif- 
ferent workmanship, since the eye really 
exacts only very general approximations to 
the mathematically correct in architectural 
ratios. But in the classic instance of the 
Parthenon it is a very difficult thesis to 
maintain that the irregularities are at- 
tributable to mason's errors or the archi- 
tect's indifference to exactness. Rather it 
appears that the builders deliberately 
sought to temper the mathematically cor- 
rect by slight departures from the "true" 
measurements, even carrying this ideal so 
far that they left no straight line straight 
nor any equal spacing equal. 8 We cannot 
fail to recall the practice of the sculptors 
contemporary to the Periclean architects, 
and particularly that reputedly Poly- 
kleitan saying concerning the "many num- 
bers" which "almost" give perfection. 
The so-called "refinements" of Periclean 



architecture, the slight deviations from per- 
fect regularity and symmetry, are not 
optical corrections for untrue illusions, 9 but 
are added in order to give lifejtojthe purely 
mathematical correctness of the eufy; 
they are the departures from the TroXAoiv 
apiOfjiwv the Trapa /xiKpoV out of which 
arises TO cv. Certainly it is a very sur- 
prising thing that the Parthenon measure- 
ments cannot be reduced to feet and dactyls 
according to any common scale. A foot of 
.2957 meters will do fairly well; a foot of 
.3362 meters will apply with nearly equal 
success; but neither these nor any other 
unitary lengths will fit all the measure- 
ments, because they are integrally incom- 
mensurable. We can only conclude that 
the builders of the Parthenon (whether by 
intelligent imitation or by intuitive artistic 
taste) had applied to architecture the same 
secret of beauty which governs natural 
forms the tempering of geometric accu- 
racy by minute deviations in the interest 
of irregularity. I need only refer to the 
extraordinary mathematical precision 
which underlies natural forms in the vege- 




table and animal worlds 10 and to the patent 

observation that in any given individual 

of the species the precision of the under- 

lying form is always tempered by the ir- 

regularities attendant upon the chance ele- 

ments of environment and growth, and add 

that we are apt to find the actual slightly 

irregularised flower or shell more real, more 

" living," and more artistically moving than 

the cold geometrical perfection of the un- 


derlying form. The form only lives when 


it is irregularised in matter. In architec- 

ture the forms are man-devised; but if they 

are harmonised with mathematical pre- 

cision and then irregularised in their ma- 

terial presentation, they will acquire a 

status analogous to that of a living thing in 


The esthetic value of using convention- 

ally established architectural forms, or 

"Orders," is again proven. For only if the 

forms are already established as species, 

can there be any chance of success for this 

device of tempered irregularity which will 

impart individual existence to each em- 

bodiment of the species-form. The em- 




ployment of the Greek "refinements" 
would be neither useful nor explicable if 
these did not constitute appreciable de- 
partures from clearly recognisable and al- 
ready familiar norms. As Greek philos- 
ophy might have stated it, the eidos is 
characterised by mathematical perfection; 
this perfection is somewhat obscured when 
the eidos is imprinted in matter; but it is 
precisely from its minute deviations and 
irregularities away from the standard form 
that the individual instance derives its in- 
dividuality and its right to a place in the 
phenomenal world of sense. 

There is a considerable body of testi- 
mony all tending to ascribe a sense of the 
unliving ("rigid," "cold," "uninteresting," 
are typical words) to buildings which dis- 
play complete precision in their measure- 
ments, and a contrary impression of life 
("alive," "warm," "elastic," "appealing," 
are frequent epithets) to the irregularised 
perfection of such buildings as the Parthe- 

It is not apparent that this principle of 
tempered precision was largely used after 




the fifth century. I believe that the frag- 

ments of later Ionic architecture show a 

very small margin of error, small enough 

to be unintentional on the part of the 

builders. The curvature of horizontal lines 

was largely abandoned, and the ideal of 

accurate perfection seems to have main- 

tained itself thereafter in most of the 

Greco-Roman as in the Renaissance and 

modern classical schools; but these state- 

ments are subject to confirmation or cor- 


* * * 


In the Ionic entablature there was no 

still at 

obvious opportunity of setting up rhythmic 

accord with the columns by a simple ratio, 

as in the Doric Order. The Ionic archi- 

trave carries no vertical ornament, the 

figured frieze is horizontally continuous, 

the crowning mouldings of the members are 

too minute to be serviceable. There re- 

main only the dentils and the sima. Of 

these, the dentils furnish a very clear ver- 

tical motive repeated in horizontal exten- 

sion; but they are rather small, and the 




consequent accord of ten to fifteen dactyls 
for each intercolumniation is not a relation 
which the eye would easily detect or from 
which it would derive any particular satis- 
faction. The sima, however, with its 
lions'-heads, could be utilised and though 
the matter has not been thoroughly inves- 
tigated it would appear that the architect 
Pythios employed these lions'-heads to es 
tablish a 3 to 1 ratio with the colonnade. 
Thus, on his great work the Mausoleum 11 
and his master work the little Athena 
temple at Priene, the lions'-heads space 
three to the column interval, and are so 
placed that two out of every triad are ex- 
actly over the eyes of the volutes of cap- 
itals as well as over the edges of the column 
shafts at their mean height. 12 By this de- 
vice the eye is led to single out at three 
levels (at mean column-height, at the 
capital, and at the roof -gutter) a horizontal 
measurement which recurs continuously, 
while every third recurrence of it is ac- 
cented by a column. With this tri-partite 
division the dentils are not brought into 
accord on the Mausoleum; but in Pythios' 




later work, the Athena temple at Priene, 
there are five dentils to each lion-spacing. 
A century later, in the Artemis temple at 
Magnesia on the Meander, Hermogenes 
seems to have adhered to this same 3 to 1 

The same ratio also obtained vertically 
in the Ionic Order; for the entablature was 
rather evenly divided into the three ele- 
ments of epistyle, frieze, and cornice, while 
the epistyle was subdivided into three 
horizontal fasciae and the cornice into three 
elements, dentils, geison, and sima, 13 
though these usually were of unequal 

Plans and facades were also laid out to 
exhibit simple ratios. The length and 
breadth of the Parthenon stylobate are as 
4 to 9. At Peiraeus in the arsenal which 
Philon built in the latter fourth century, 
the front, beneath the cornice, was a per- 
fect rectangle, twice as wide as it was high; 
on the long sides the ratio of length to 
height was exactly 15 to 1. The interior 
floor space was a rectangle whose sides were 
related as 8 to 1. 




Finally, attention might be called to the 
predilection for "round" or perfect num- 
bers. Athena's old temple on the Acro- 
polis was known as the "Hundredfoot " and 
the sanctuary room of the Parthenon may 
perhaps have kept this measurement for 
its length. The cella of the Athena temple 
at Priene measured 100 feet in length from 
wall-face to wall-face. 

The use of Nnmber penetrated into ar- 
chitectura Ipractice much farther than the 
mere horizontal rhythm or the simple com- 
mensurability of the larger elements. In 
the fully established canons of the fourth 
and third centuries, in the work of the 
Ionic architects Pythios and Hermogenes, 
it would appear that every smallest ele- 
ment was dominated and determined in its 
measurements by numerical relations. In 
trying to unravel their skein of numbers, 
the student will discover that much of the 
commensurability may be ascribed to the 
architect's use of a foot-rule, and will be 
tempted to believe that this simple ex- 
planation is sufficient. Where stones were 
laid off to a measure of feet and dactyls 




they could not well fail to be commensur- 
able. But since there were 16 dactyls to 
the ancient foot, divisions and subdivisions 
would have to partake of the nature of 
continued bisection: division by 3, 5, or 
7 would seldom yield lengths measurable 
in dactyls. Yet if Dinsmoor's measure- 
ments of the Mausoleum fragments in the 
British Museum are correct, such com- 
mensurable but not measurable 14 relations 
were employed, and Vitruvius refers fre- 
quently to division into 7, 9, 11, 13, etc. 
It would seem therefore that the com- 
mensurability of parts is not an accidental 
corollary of the employment of a measured 
foot-rule, but an intentional artistic prac- 

A further objection can be raised. Vi- 
truvius, who must have had access to the 
traditional canons of some of these Greek 
architects (Hermogenes in particular) is 
full of rules of perfection. He instils com- 
mensurability (modulatio or commodulatio) 
by determining a modulus or unit with 
which to measure off diameters and heights 
and widths. Out of any measurement so 




established a new modulus may be derived 

for fixing details and lesser elements in a 


Thus, for example, the heights of 

mouldings are stated as fractions of the 

members to which they belong; the mid- 

dle fascia of an Ionic epistyle is taken as 

a modulus for the geison; the diameter 

of the oculus of an Ionic capital gives 

the amount of projection for the echinus, 

and so on. This method of passing from 

one modulus to another is nowhere more 

clearly expressed by Vitruvius than in 

his description of the Ionic doorway. 

. . . From this example we see that 

though each member of the doorway is 

regarded as a modulus or measure of its 

immediate neighbour, nevertheless all 

are connected with each other and with 

the large dimension of the whole by a 

common measure. This illustrates the 

Vitruvian conception of proportion and 

there is every reason to believe that the 

standpoint of the Greek authors from 

whom he 'derived his inspiration was not 

essentially different. 15 





It is a frequent impression among modern 

readers and commentators of Vitruvius 

that these rules for the minutest details of 

the elements of the Orders are intended to 

be purely practical devices to save time and 

trouble with simple rules-of -thumb; and 

so they may have been in Vitruvius' own 

time and since then. But all our discus- 

sion has tended to show that this is not the 

whole explanation, but that we have in 

these canons the echo of that old faith in 

the efficacy of Number to establish the per- 

fection of the form. 

* * * 





I have dwelt a good deal on this nu- 
merical aspect of Greek architecture not 
only because it is philosophically interest- 
ing in its own right, but also because of 
certain remarkable implications which it 
involves. Out of Number come com- 
mensurability of lengths and surfaces and, 
by repetition, architectural beat and 
rhythm. But all these in Greek archi- 
tecture are only effective upon the spec- 
tator if the matter in which they are em- 
bodied is seen as in one and the same plane. 

The rhythm of a Doric colonnade with 
its measured recurrence of columns moving 
past, while the triglyphs double and the 
cornice and roof quadruple the same steady 
beat all this is effective, indeed exists for 
the beholder, only if all these elements are 
felt to be on a single surface or area or 
plane. Theirs is an art of related lines and 
surfaces, not of solids. 

Now it is perfectly true that "inequality 
is the normal fact of optical appearance" 16 
and that none of the equal surfaces of suc- 

Spatial pre- 
once more. 
(Cf . chap- 
ter II.) 
Number a 
function of 
sional pre- 






ace ordingly 
an art of 

wo dimen- 

cessive metopes or the equal spacings of 
triglyphs or regulae or lions'-heads will 
really be equal in our field of vision be- 
cause of the perspective in which they are 
seen. But the mind has learned to cor- 
rect the appearances which the eye en- 
counters and to read off the true state be- 
hind the changes of perspective vision. 
But it does this, in Greek architecture, by 
the assumption of a plane in which all the 
unequal appearances actually lie with all 
their appropriate lengths and areas equal. 
It is by referring them to a single plane that 
the mind reconstitutes that recurrence of 
equal intervals on which the sense of 
rhythm depends. 

It should follow that Greek architecture, 
n order to make its use of number effective, 
must have been an architecture of planes 
rather than of solids. This was precisely 
the case. 

Instead of true depth, Greek architec- 
ture gives us successive and usually parallel 
planes. The stoa or colonnade presents the 
ront plane of the exterior order, the middle 
plane of the columns which support the 




roof-beam, and the rear plane of the wall 
at the back. There is no sense of depth or 
enclosed space. Rather, the true depth 
has been converted into a succession of 
nearly flat and wholly disconnected sur- 
faces precisely as in Greek relief sculpture. 
Even the roof, thanks to the straight 
mounting lines of the cover-tiles, joins in 
the vertical planes. The human beings 
moving in this colonnade appear to the ex- 
terior spectator like bas-reliefs on the rear 

With the temple it is, of course, the 
same. On the long sides there are two par- 
allel planes, that of the colonnade and that 
of the naos wall; on the short sides there 
are three planes, that of the exterior colon- 
nade, that of the vestibule colonnade and 
that of the rear wall of the vestibule (with 
or without its doorway with door or hang- 
ing). More complicated vistas in agora or 
temenos only add other planes. In 
glimpses across angles of colonnaded 
squares, or through corners of dipteral 
temples, the masses are all presented at in- 
tervals so that the eye sees a series of 




superimposed surfaces sharply divided by 
an intervening break of perfectly indeter- 
minate extent. It should be added that 
the Greek atmosphere encourages this im- 
pression by eating up the air's appearance 
of solidity, so that space loses its density 
and surfaces seem directly superimposed. 

Not all these planes are parallel since 
the colonnades surround the temple on the 
exterior and frequently the agora or te- 
menos on its inner face. But only one angle 
is permitted, the right angle of 90 degrees. 
Just as the mind interprets the perspective 
of the stoa by its knowledge that it lies in 
a single plane, so it can proceed to the more 
difficult perspective of a corner if the angle 
is known and constant. The rectangular- 
ity of Greek plans is their most remarkable 
characteristic. Temples, temple enclos- 
ures, market-places, houses, and (after the 
fifth century) towns and cities are rect- 
angular without break or pity, as the 
architecture of planes demanded that they 
must be. 

And so there follows the most remarkable 
property of all in Greek architecture. 



Since its appearances all lie in parallel or 
perpendicular planes, it can only define or 
bound solid space, and cannot enclose it. 

Just as there is no true depth to a Greek 
colonnade, but only the specious depth of 
reek sculptured relief, so there is no sense 
of space shut in and contained in anything 
that the Greek builders made. Even the 
temple-interiors offered only more colon- 
nades and walls in parallel planes, topped 
by a flat ceiling which was only another 
plane at right angles. Such a procedure 
bounds space, defines it; but though me- 
chanically it contains space because it 
shuts it off, it is powerless to impart any 
such feeling to the human mind. Whereas 
a vault closes down upon space and makes 
space sensible, a panel ceiling merely ties 
the walls. In the terms of relief-carving 
it acts like the straight edge or step be- 
tween successive planes. 

That these temple interiors were but 
half-lighted is perhaps an indication that 
artistically they were but half -felt or half- 
considered. They were but the space en- 
closed by the inner faces of the walls; they 





were the inside of the treasure-chest, serv- 
ing to shut in the gold and the sacred heir- 
looms; and just as the outside of a chest 
has the craftsman's favor and gets the 
carving and gilding, so it was the exterior 
aspect of these enclosing walls which ap- 
pealed to the ancient builder. 

No doubt this is somewhat fanciful, and 
there exists a better reason for the opinion 
that interiors were the least successful 
aspect of Greek architecture. For it is of 
the utmost importance to note that Greek 
architecture had almost no need (of a di- 
rect practical kind) for trying to shut space 
in. Greek life was out of doors. The 
houses were mere sleeping-cells about a 
central patio, the theatres were unroofed 
gathering-places of the people upon the 
hill-side; the market-places were open; 
even the shops were mere store-houses from 
which goods for the day's trade could be 
brought out to be spread in the open 
bazaar; the colonnades were but casual 
out-of-door shelters from wind and rain 
and sun; the schools were out-of-door 
palaestrae; the hospitals were colonnades 



in the sacred precinct of Aesclepios. Only 
the temples and the council-rooms so much 
as put the problem of enclosing air-space 
with encompassing masonry. Before the 
late Hellenistic age, domestic architecture 
hardly existed. It is not remarkable, 
therefore, that the architects devoted 
themselves wholly to external appearances 
or that, given the Greek conditions of at- 
mosphere and light, they learned, as no 
other race has done, the secret of surface 
and line, but got no farther in the esthetics 
of their art than the expression of support 
in vertical parallel planes. 

We may assert, therefore, that the Greek 
architects had no comprehension of the 
artistic handling of space in interiors. 
That vast preoccupation of Gothic and 
modern designers was to them a book un- 
opened. If a paradox is permissible, their 
only interior compositions were out-of- 
doors, where the roof was the blue sky. 
In town-square and temple-precinct they 
composed spatial complexes; but the pic- 
ture is always made of overlapping sur- 
faces moving across one another as the 




spectator moves. The intervening depths 

are lost, just as they were lost for the 

ridges and gulleys and forests of their own 

beloved mountains which still to-day show 

like flat bas-reliefs against the Grecian sky. 

* * * 


Architecture makes its esthetic appeal 


visually, utilising both two-dimensional 

tions of 

and three-dimensional presentations of 


itself to its beholder. For two-dimensional 

presentations it depends upon its appear- 

ance in the flat, as though it aimed at plane 

composition, with a design in areas whose 

mutually related shapes and sizes, colors 

and positions were the object of esthetic 

contemplation and the source of esthetic 

delight. Almost any facade is a composi- 

tion of this kind. The Italian Renaissance 

carried to a high refinement and subtlety 

the art of such planilinear presentation. 

The fronts of the Municipio at Verona and 

the Grimani palace in Venice are instances 

of compositions for whose appreciation we 

shall need some feeling for the alternating 

rhythm with which wall-space and window- 





space, light and shadow, column and col- 
umn, are held against each other for con- 
trast of recurrent agreement. 

Three-dimensional presentation may be 
most intelligibly explained as the effort of 
architecture to suggest the enclosing of 
space. In sculpture we saw art striving to 
impart a sense of space solidly occupied. 
In architecture we have a striving to out- 
line, define, and limit space within an en- 
closing shell or boundary. It is not enough 
that walls and roof do actually enclose 
space; it is the artist's part to make them 
seem to do so, that the spectator may by 
direct visual apprehension be aware of this 
power and property of the builded thing. 
St. Sophia, seen from without for all its 
clumsy contours most patently succeeds 
in imparting this sense of space enclosed. 
To do so is indeed one of the original and 
distinctive qualities of the Byzantine style 
which differentiates it from the preceding 
Greek as thoroughly in the esthetic domain 
as the use of domes on pendentives marks 
it off in the mechanical domain. It is ob- 
vious that since architecture, unlike sculp- 


and three- 




ture, is intended to be seen from within as 
well as from without, the artistic problem 
occurs not merely for the visual presenta- 
tion of the exterior but recurs even more 
insistently for the interior. And here, 
since the spectator is himself physically 
within the enclosed space, the three-dimen- 
sional presentation is of the utmost impor- 
tance and carries with it an extraordinary 
range of emotional affection. The spec- 
tator is peculiarly at the mercy of every 
spatial suggestion, so that an almost mag- 
ical power accrues to the intelligent master 
of this art. The apprehended space may 
bear almost no similarity to the actual 
physical space of such-and-such cubic 
volume which is determinable by measure- 
ment. One has but to recall the nave of a 
large Roman basilica and that of a major 

othic cathedral to understand how little 
propos are the actual dimensions of 
tieight and width. 

It is just this power of three-dimensional 
presentation (so marvellous an achieve- 
ment of the Gothic style) which I feel to be 
almost entirely absent in ancient Greek 




architecture. It confines itself to a two- 
dimensional presentation. 

Indeed, I am tempted to hazard the 
seeming absurdity that the East Mediter- 
ranean people live in a much less three- 
dimensional world than the North Euro- 
peans. A visual space in which all distant 
objects tend to be projected upon a single 
plane would not present so much depth, so 
much three-dimensional solidity, as a vis- 
ual space in which the correct distance of 
objects was more clearly defined. A race 
living in a spatial environment of the first 
type would incline in its art to silhouette 
drawing and flat "decorative" design. It 
is not accidental that Egyptian drawing 
never succeeded in indicating aerial depth 
and distance, or that a great power for 
enveloping painted scenery in atmospheric 
shadow arose in the school of the Nether- 
lands. In Egypt and Arabia the shadow 
of a modelled projection tends to look like 
a dark surface in the same plane as its back- 
ground and it is therefore disturbing to the 
beauty of the design unless its superficial 
area and outline agree with the rest of the 

(but not in 


and space 



pattern. How much of the quality of 
pierced black-and-white ornament of 
Mschatta, Syrian Byzantine, and Coptic 
decoration is not explained by this at- 
mospheric accident? And how much is 
not the Perso-Mesopotamian genius for 
decorative pattern in rugs and hangings 
and fayence a function of an atmospheric 
compulsion toward flat design and toward 
polychromy as a substitute for the chia- 
roscuro of varying projections and model- 
ling? Conversely, in the rainy and foggy 
lands, third-dimensional depth is directly 
given in visual perception, since projec- 
tions and shadows keep to their own plane 
and hold their distance. The repeated 
mouldings of Gothic portals and piers de- 
pend greatly upon this possibility of real 
spatial extension of objects in light and 
shadow. Is it any wonder that the deep 
portals and the great Gothic interiors never 
grew in Italian but flourished in Northern 
European air? And is it not a futility to 
take Greek architecture, calculated for 
visual projection upon parallel planes 
which the climate of Greece enforces, and 




to transplant it to Liverpool or Berlin where 
it will be riddled with depth, like a curtain 
shot through with holes? 

Upon this two-dimensional character of 
its visual presentation Greek architecture 
based its artistic effects. Since support in 
the immediate vertical plane was all that 
such an art could present to the eye, all 
expression of weight sustained and held 
aloft must appear in the arrangement of 
lines and surfaces in the frontal plane of 
the Order. The entablature presses flat 
upon the columns and therefore can toler- 
ate only horizontal or vertical lines in its 
decoration, since in this way the heavy un- 
comprising mass can be made directly in- 
telligible. Our eyes follow the horizontal 
lines and our emotions instinctively in- 
terpret such lines in terms of mass in a state 
of rest. We have experienced just such a 
sweep of line in looking at the horizon of 
plains or of the ocean: it is the line of things 
outspread, heavy and immovable and en- 
during, held by gravity, but in no danger 
of falling. The columns express in their 
contour successful but not effortless re- 


of Greek 
ture in its 





sistance to this downward pressure, which 
weighs squarely and heavily upon them 
until they bulge with the pressure. Here 
presumably is the psychology of entasis: 
it is the visible yielding to compression 
until a stable condition of resistance is 
reached. It establishes an analogy with 
human muscular effort and presents it 
directly to our emotional sensibility. 

Obviously the Doric Order with its heav- 
ier and simpler entablature, more massive 
columns, and more pronounced entasis 
carries this appeal much farther than does 
the Ionic, which has comparatively a light 
and variegated load which it carries easily 
and with elegance. It is because of this 
direct visual presentation of the vertical 
support of a great weight, not without ef- 
fort, that the Doric temples are so " vigor- 
ous," so "powerful," so "enduring" and 
"eternal." The nearby rocks against 
which the sheep rub their flanks have lasted 
in their places longer than these ruined 
temples; but it is the temples which cry 
aloud that they are everlasting. 

As long as the mere problem of overcom- 



ing mechanical difficulties besets and trou- 
bles the builder, he will take delight in 
visualising through architectural means the 
weight of stones and the effort of holding 
them aloft. This is why so much early 
work of the human races is megalithic or 
massive or (where the race is artistically 
more sensitive and more gifted) more im- 
posing and more able to impart a sense of 
great weight nobly borne. But when the 
mere mechanical obstacles cease to be 
obstacles, the builder loses his feeling for 
the sheer oppression of weight and tends 
more and more to delight in imparting the 
very opposite sensation of stone poised 
aloft as though it weighed nothing, and 
using it for a free fantasy of imaginative 
constructions. The analogy between Ro- 
manesque with the successive three periods 
of Gothic and the progress of Greek archi- 
tecture is therefore not casual but neces- 
sary. The Doric Order is gradually robbed 
of its sense of size and weight and the Ionic 
Order grows more and more universally 
popular only to yield to a Corinthian Style 


A sugges- 




which grows (before its final decadence) 

extremely ornate. 

The need for fluting the column is easily 

explained. Seen in its out-of-door setting 

in the vertical plane of its Order, the Greek 

column would appear as a thin, flat surface 

without sufficient solidity to perform its 

function of support, were it not for the 

flutes, the peculiar spacing of whose ver- 

tical shadow-lines forces the eye to inter- 

pret the column as a rounded solid, how- 

ever invisible the curvature may otherwise 


As in Greek relief, the third dimension 

is not suppressed, but abbreviated. It 

exists to give substantiality to the content 

of each of the various planes. 




It is often said of Greek architecture that 
it has no method of suggesting size because 
the members of the Orders maintain such 
accurate relative proportion that there is 
no fixing of "scale." So far was this accu- 
racy of proportion carried that correction 
was made for the apparent diminution of 
size of distant objects by making them cor- 
respondingly larger than strict ratio de- 
manded, as when entablatures topping high 
columns were given more than their strict 
proportional height. (Cf. Vitruvius III, 
3, 13.) Consequently, by eliminating even 
that clue to actual size which the diminu- 
tion due to actual distance from the eye 
might have furnished, the possibility of 
judging scale was wholly removed. This 
is certainly true; but it is also true that if 
any absolute unit of measurement could 
be introduced, an apprehension of the size 
of every part would follow, precisely be- 
cause of this constant accuracy of relations 
between the elements. If it is recalled that 
Greek architecture was intended for public 






Contour " 

places, it will be evident that the Greek 
colonnade was usually seen with human 
beings standing or moving among its col- 
umns. The seen relation of column-spac- 
ing to the width of men's bodies, and of 
columns' height to the height of men, sup- 
plies just that absolute unit of measurement 
on which scale depends. We may under- 
stand from this why Greek taste tolerated 
the fixing of statues between the columns 
on the top step of a temple platform (as at 
Olympia) or the amassing of statuary in 
open agora and temenos. Just because of 
the orderliness of Greek ratios a human 
being or a life-sized statue could furnish 
scale to a degree wholly impossible in other 
periods and styles. 

Greek relief, we have seen, was an art 
mainly of contours. If the insistence on 
the two-dimensional quality of Greek archi- 
tectural composition is justified, it should 
bllow that the Greek architects should 
lave been extremely sensitive to outline; 
and this is notoriously the case. The cor- 
ner of a Grecian temple seen against a 
uminous sky should be a revelation to 



every modern student. The broken line 
from step to akroterion is amazingly clari- 
fying, in that it so clearly distinguishes 
and emphasises every member of the Order. 
More than that, it helps to set the mass of 
the temple in equilibrium and stabilizes the 
whole structure. But again it is essentially 
as the bounding-line of a plane surface that 
this corner contour works on us as one 
can see from the extraordinary way in 
which its effect is heightened at night or 
against a sunset sky, when the temple loses 
its projection and solidity and flattens to a 
surface in silhouette. 

Since its effects are calculated for pres- 
entation in a plane, what should have been 
the fate of the classic Orders in later times 
when the great discovery of architecture as 
an art of three-dimensional presentation 
had been made? They should have sunk 
to an ornamental accessory for articulating 
facades and other vertical planes. Actu- 
ally, as early as imperial Roman times, they 
began to be used as attached orders with- 
out structural importance; and though 
purely Hellenic structures have continued 




to be built, it is largely as decorative 

motives that the forms of Greek architec- 

ture persist to-day. Their fate was thor- 

oughly appropriate and deserved. 

An un- 

Like Greek painting and Greek music, 17 
Greek architecture marks a stage of in- 


complete development in the evolution of 

its art and, like them, it exhibits the most 

enviable and amazing perfection of certain 

cardinal esthetic elements, combined with 

a complete ignorance of others no less im- 

portant to the art in its fuller evolution. 

* * * 





Where mistakes are costly and produc- 
tion tedious, patrons and artists alike step 
warily. Printer's ink and paper, canvas 
and oils can be lavished, their cost is small 
restraint; but the hewn stones, the scaf- 
folds, and the vast array of masons make 
architectural follies too prodigal for repeti- 
tion. For this reason, though we may write 
for amusement and paint for pleasure, we 
seldom build except for use. Finally, 
though poetry can turn the hills upside 
down, live in the moon, and walk on thin 
crusts of water, though painting can re- 
make the world to suit itself, architecture 
must rest a stone on a stone, even to span 
a void, and desist when the formula of 
gravitation says No. 

Here, then, are three considerations 
which keep architecture within conven- 
tional rounds: cost and difficulty of pro- 
duction, practical aim, physical limita- 
tions. Because the claims of these three 
must first be satisfied, they have made 
tyrants of themselves and have even gone 


ture as an 
art of visual 




who must 
be over- 



so far as to elevate their demands into a 
canon of architectural beauty. 

Cost has declared: "A building is 
beautiful when no element of it is super- 
fluous." Practical Aim has asserted: "A 
building is beautiful when it expresses the 
use for which it was made." Physical 
Limitation has maintained: "A building 
is beautiful when it expresses its construc- 

But all these are mere tyranny of mate- 
rialism and have little or nothing to do 
with artistic values. At most, their truth 
is purely negative, for it may be that if a 
building does not express its use and pur- 
pose, the offense against our practical sense 
produces an esthetic reaction, and if a 
building looks as if it were supported by 
columns and beams, while really it rests on 
concealed vaulting, our like of straightfor- 
ward dealing and our displeasure at decep- 
tion may influence our esthetic judgment. 

For years I have tried to persuade my- 
self that my pleasure in a Gothic cathedral 
was derived from an appreciation of the 
marvellous balance of thrusts and strains, 




that I approved of Roman aqueducts be- 
cause they were nothing but arches reaching 
up to a level at which they could make 
water flow across plain and valley to a 
far-off town, and that the plan of a Greek 
temple was a perfect thing because it 
showed a room for the god's image, a room 
for the adorant, a preparatory vestibule, 
and a surrounding portico with shelter 
from sun and rain just the things that 
were needful, and nothing more. And all 
the while secretly I knew that I was per- 
suading myself with an intellectual theory 
and that the whole thing was simple self- 
deception, that I was taking a contributory 
element for sole cause and criterion, and 
that I liked Gothic cathedrals and Roman 
aqueducts and Greek temples for much 
more simple and less round-about reasons, 
that I liked them not because they ex- 
pressed this or that, but because their 
looks appealed directly to me. I liked them 
not merely for their intentions, but for 
what they were so much stone in such and 
such shapes. In fact, just as I liked music 



before we 
can pass on 


for what I heard, I liked architecture for 
what I saw. 

I suspect that most of us must pass 
through some similar experience before we 
are prepared to accept any of the theories 
now to be developed. As long as we are 
persuaded that the esthetic of architecture 
is based on materialistic tyrannies such as 
utility and economy, we relegate archi- 
tecture to a different category from that of 
the other arts. No doubt, an architect 
who does not practise economy or empha- 
sise utility will run the risk of being 
rightly refused employment. But this is 
merely to say that an architect should be 
a good engineer, and is not at all the same 
thing as to maintain that engineering and 
architecture are one and identical. I 
should like to see engineering defined as the 
science, and architecture as the art, of con- 
struction; because, if once we can dismiss 
the science by fully allowing its necessity 
and importance, we shall have an oppor- 
tunity to discover what the art is. 

Engineering was a vital matter to the 
builders of the pyramids, who seem to have 




mastered its elements so well that their 
mechanical adroitness has rather puzzled 
succeeding generations. But does that 
fact give us any hint of the real reason 
why the pyramids impress us as they do? 
Surely the primary reason is because a 
simple and elementary shape is presented 
in a gigantic size. And that is all there is 
to it. The rest the difficulties of con- 
struction, the nearly 5000 years through 
which they have endured, the desert sand 
on which they rise, our imaginative and 
historical impressions of Pharaoh-ruled 
ancient Egypt, our literary enthusiasm for 
Cleopatra who had nothing whatever to 
do with the pyramids all contribute, 
joining and fusing themselves to a complex 
emotional reaction; but they are not the 
fundamental of the architectural emotion, 
to which, indeed, they are in one sense 
thoroughly irrelevant. 

In late Gothic fan-vaulting, the visible 
elements express a structure which is not 
merely physically absent, but is physically 
impossible. The ceiling in its periphery 
rests upon the ribs, the ribs rest upon the 




central ornament, and this in turn rests 
upon nothing! Were the forms clumsy, 
the lines oppressive, or disconnected, the 
spectator's emotion, I imagine, would be 
one of terror and fear; but because the 
lines are stable and buoyant and strive up- 
ward, our emotion is one of delight. Mass 
has lost its oppression; it floats upward like 
the Walhalla architecture of the thunder- 
cloud. We know it cannot be, yet we see 
that it is. Is it not captious and pedantic 
to complain that Henry VII 's Chapel is 
not good architecture because it violates 
a fundamental principle of the art? Rather 
we should reconsider the validity of our 
assumed principle in the face of such a re- 
futation of its authority. Again, what of 
the dome of St. Sophia that does not seem 
to come down with any weight upon its 
four pendentives? Let us not insist that 
good architecture must express its construc- 
tion. Let us say that to express structure 
is one of architecture's sources of appeal, 
and that good architecture accordingly ex- 
presses structure vividly, immediately, 
appealingly, stimulatingly, what you will, 




if only you will leave the art free rein to do 
what it chooses with its own esthetic de- 

What of the facade of St. Mark's in 
Venice? If we removed everything that 
did not contribute to constructional neces- 
sity or practical use, I have a very shrewd 
idea of the result; for the structure is 
based on pre-Gothic methods of building 
while the ornamentation is Gothic. Is 
there not something of blind obstinacy in 
persisting that the Gothic ornamentation 
is therefore an unpleasant accretion, be- 
cause it is structurally irrelevant? 

Or of what use in architecture is colour, 
except when it helps to emphasise struc- 
ture and utility? None; therefore it is 
really irrelevant? Architecture calls in the 
arts of design to make her finished work 
more pleasing? They are accessory, not 
an integral part of architecture? Blind 
obstinacy again. Because a building will 
stand as well, in whatever way its walls be 
"decorated"; because a green arch. will 
last as long as a red one; because a mosaic 
is in the surface, and not in the mechanics 




A definition 


of support; because colour can be put on 
after the work of building is ended; be- 
cause of a host of reasons such as these, 
colour is dubbed irrelevant to architecture. 
True, since colour and construction are in 
different categories, colour may be wholly 
irrelevant to engineering; but why irrele- 
vant to architecture? why accessory just 
because, in the making, it is usually sub- 

May there not be a more purely esthetic 
aspect of architecture, in which it appears 
to us as an emotional art in three dimen- 
sions which employs pure forms in a visual 
appeal, working on our susceptibilities of 
mass, outline, colour, and pattern, our mus- 
cular sense of balance, of strain, of freedom 
of motion and confinement, of size and 
weight and power, and embodying all these 
in a construction whose right to existence, 
because of its human utility and structural 
sanity, our reason approves? 

By such a definition I refer to nothing 
abstruse or over-sophisticated. A high, 
black wall projecting above us in a narrow 
street with squat overhanging masses is 




terrifying even when our reason assures us 
that it cannot fall. A colonnaded street 
like that of Palmyra, with its perspective 
narrowing toward a vanishing point, gives 
us a sense of reach and distance for which 
the actual linear extension seems no ade- 
quate warrant. The vertical lines of rib- 
bing and grooving in a Gothic pier give an 
immediate realisation of height which a 
smooth shaft would not impart. A great 
space easily crowned, like St. Sophia with 
its dome on pendentives, gives us a sense 
of freedom to move; while the hall of a 
thousand and one columns underground, 
the Ben-bir-direk, makes a nightmare of 
the thought of bodily movement through 
space. It is with a pleasurable confusion 
that the eye loses itself in the weltered 
glory of Gothic rose-windows filled with 
stained glass, and an unconfused self- 
certainty with which we greet the picture- 
less sharp outlines of a Greek temple 
against the sky. These are simple, almost 
grossly simple, examples of emotions to 
which we all yield, because they are not 
supplied by us, but to us, being inherent in 




yet little- 

the very lines, surfaces, bulk, and colour of 
the things at which we are looking. Here, 
as in other arts, pure form is at work, giv- 
ing us the primary emotions to which all 
our sophications of historical, mathemat- 
ical, cultural, and critical appreciations or 

associations are accessory. 





What then are these pure forms of archi- 

All the forms for painting and sculpture 
seem to belong to architecture as well; but 
because they are not so obviously made to 
fuse with a represented object, they cannot 
appeal to us so strongly. They remain 
more nearly in a state of pure form con- 
crete enough, since they are exemplified in 
stone or wood or plaster, but not merged 
in something other than themselves. In 
lieu of imitations and illusions of real ob- 
jects on which these forms are overlaid, 
architecture shows us only certain con- 
ventional entities simple elements such as 
columns, capitals, triglyphs, piers, groins, 
dosserets, or composite elements such as 
porticoes and blind-stories. Perhaps as 
was suggested previously we can here dis- 
cern a reason why architectural tradition 
clings so tenaciously to its own artificial 
creations, such as its five classic Orders with 
all their rigid specifications of proportion 
and detail. For if columns and capitals 


An analysis 
such as we 

applied to 
the other 
arts is the 
here also. 
It is at- 
tempted in 
the follow- 
ing pages 
and there- 
with brings 
to an end 
our study 
of the 
Greek arts 




must be shaped just so, and not otherwise, 
they acquire something of the definite in- 
dividual reality of a product of nature. 
A column seems to be a column, almost as 
a dog is a dog or a tree is a tree. The archi- 
tectural elements become part of the world 
of our ordinary experience, There may 
therefore be a closer parallel than we should 
have thought between a painted pattern 
expressing itself on a landscape and an 
architectural pattern expressing itself on a 
classic structure. Both landscape and 
classic Orders are repeated from the famil- 
iar forms of our visual experience. And 
the parallel with sculpture is even closer. 
Where the sculptor imitates living objects, 
the architect imitates the objects of a tra- 
ditional world of recognised shapes. Both 
sculpture and architecture are representing 
something other than what (materially 
considered) they actually are. 

In saying that architecture employs the 
pure forms of painting and sculpture, I of 
course am not referring to the familiar fact 
that architecture uses these arts for dec- 
oration. I mean that a building's fagade 




may exhibit pattern, balance of masses, 
visual guiding lines, just as a painting 
might do, and employ flow or break of 
outline, suggestion of weight and resistance 
and balanced strain, just as a piece of 
sculpture might. 

But in addition to these now familiar 
forms, architecture has her specific oppor- 
tunities to work upon our emotions. We 
are always external to sculpture, but the 
works of architecture enclose us spatially, 
hem us in and ring us round with vast 
masses of stone many times our height and 
weight. Not only do we crane our heads, 
focus our eyes for distance, perform actual 
physical exertion in moving from point to 
point; but we are at the mercy of sugges- 
tions of confinement, freedom of movement, 
oppression, physical danger. Our eyes 
travel up with us easily to vast heights, or 
struggle hopelessly over the horizontal 
barriers which keep us down. We have 
vast cisterns of air to breathe, or our lungs 
gasp under the illusion that we are closely 
shut in. There is, indeed, an entire range 
of bodily experiences, for the most part not 




very prominent in our consciousness, to 
which the formal suggestions of architec- 
ture may appeal. Such are our sense of 
poise and bodily balance, of muscular self- 
control, accuracy of movement, lightness 
and agility, feelings of strength and self- 
assurance, freedom of breath, even such 
vague bodily states as accompany security 
of footing, indifference to external forces, 
determination and endurance, boldness and 

Thus analysed and put into words, the 
sensations of which I have been speaking 
as aroused in us by architectural forms 
seem artificial and unreal. Actually they 
lie so lightly on us that qua physiological 
conditions in us we are seldom even con- 
scious of their existence, but tend to ascribe 
them as actual qualities to the architec- 
tural object. But I do not wish to insist 
on the psychological or physiological mech- 
anism of artistic emotion. The essential 
thing for my purpose is the recognition 
that what I have called artistic form ap- 
peals directly to certain sensibilities in us, 
and leads us to ascribe to the work of art 



emotional values with which we should 
otherwise have no acquaintance. That 
these emotions or qualities are in the build- 
ing is precisely the opposite of my meaning. 
These things are in us, and it is to our 
sense of them that architectural form can 
appeal. When we exclaim at the lightness 
of crocket and finial, there is no gravita- 
tional lightness in the carven stone. We 
are apt to speak as though tracery were 
done in pumice and wall-bases in solid 
ore. A chisel may remove only a few 
ounces from the actual weight of a stone, 
yet the stone may thereby become so light 
that it almost floats in air from its own 
buoyancy; for the sense of its lightness is 
in ourselves. The common sense behind 
traditional ways of speech bears testimony 
that these are familiar human experiences 
and not mere sophistications. 

But we ascribe the lightness to the stone. 
The appeal is to our own sense of equilib- 
rium or freedom of movement or whatso- 
ever it may be; but what we feel in our- 
selves we read into the objects which we 
are contemplating. Just as, in painting, 




our trivial actions are fathered upon an 
imaginary spatial world and thereby ac 
quire importance and magnitude, so in 
architecture our reactions, of no moment 
in themselves, when they serve to vitalise 
a towering mass of stable stone, arouse 
emotions that can overwhelm us by their 
scale and power. 

How intentionally do architects employ 
these emotional forms? That must depend 
on the architect and the period. The 
great masters of the Renaissance and, most 
of all, of the succeeding " decadence" or 
Baroque period, made a far-reaching study 
of these forms (though they did not so 
call them, or so philosophise about them). 
The restlessness, the turgid pomp, the 
training after greatness which charac- 
-erised the Baroque are not accidents or in- 
idents of stylistic evolution, but inten- 
ional emotional qualities. Nineteenth 
century criticism, with its materialistic 
and "scientific" penchant, saw in the 
rolutes and scrolls and urns and super- 
mposed pediments and the "clumsy" 
masses of many an Italian Baroque fagade 



only an unnecessary waste of material with 
irrelevant ornaments run wild and obscur- 
ing structural design. But that is to judge 
from parti pris, to beg the question; for it 
assumes that the aim of good architecture 
is to express structure and purpose with 
economy, whereas the Baroque architects 
were trying to express something quite dif- 
ferent. They were not trying to call atten- 
tion to questions of construction and engi- 
neering, but to work upon the human emo- 
tion of the passers-by. 

There are other critics who complain 
that these emotional effects are melo- 
dramatic, that Baroque greatness is mere 
bombast and pretentiousness, that the in- 
sistence on effect is tedious and out of place 
in a monumental art. This is a very dif- 
ferent angle of attack. It appeals to the 
canon of taste and with this I have no in- 
tention of taking issue; for I do not wish to 
ask what emotions architecture ought to 
evoke, but merely to consider what emo- 
tions architecture does and can evoke and 
how this is done. For that purpose the 
Baroque is a very good period to study, 18 




because it is not interested in monumental 
calm but in the excitation of emotions 
whose presence and character are easy to 

We are awed by the contemplation of 
mere bulk and size; for we make the com- 
parison with ourselves and realise our own 
ineffectiveness. The contrast with our 
own size and weight is essential to this re- 
sult. Hence it is architecture's business, 
if she wishes to produce the emotion, to see 
to it that we shall make the comparison. 
She must give us "scale," must furnish us 
with a unit whose size relative to ourselves 
we know and can appreciate, and make this 
unit operative as a unit of measurement 
which we are induced to lay off and repeat 
until we become aware how many times it 
is contained in the architectural surround- 
ings. If that repetition appear intermin- 
able or confusedly great, the impression of 
size will be all the stronger. 

Artaxerxes Mnemon built himself a 
colossal throne-room in his palace at Susa. 
Its appearance of size was not derived 
solely from the great linear dimensions of 



the floor and walls, but from the persistent 
repetition of great columns whose verti- 
cality was emphasized by the grooves which 
ran unbroken from the bases to the great 
bull-headed capitals bearing the ceiling- 
beams. Wherever one stood, one of these 
columns rose beside him and stretched 
away above him, so that a direct com- 
parison between human stature and the 
loftiness of the room was inevitable. 
Round about on all sides were these same 
columns, like the trees of a forest shutting 
in the view, endless vertical lines, all the 
same, and all enormous. Here was great 
height immediately to be apprehended, 
and endlessly and bewilderingly repeated. 
Truly, great would have seemed the king 
who throned in such a room and com- 
manded its architects and builders. 

The interior of the nave of St. Peter's 
in Rome gives no such sense of vastness. 
The classic Order there shows the same 
elements, and the same relations between 
those elements, that we might see in many 
other buildings. Standing in the open 
floor-space, we have little from which to 




measure size in relation to ourselves. What 
unit we may derive, is repeated in a quiet 
and ordered way for a certain number of 
times. With no immediate apprehension of 
size, and no uncounted repetition of great- 
ness, we have little but our reason to tell 
us that we are in a hall of vast extent and 
height. ' 

I have taken sense of size as an example, 
because it is easy to see that it depends not 
wholly on actual dimensions but on some- 
thing in the arrangement and distribution 
of the architectural elements. It is the 
readiest and most persuasive instance of 
the effect which a purely formal principle 
may have upon our attitude toward a 
building. But it is easy to convince our- 
selves that formal laws play a very ex- 
tensive rdle in our most every-day impres- 
sions of architecture. Buildings are "top- 
heavy," "crooked," "forbidding," "at- 
tractive," "sombre," "full of energy," 
"magnificent," "stiff" (to use only a lay- 
man's vocabulary and a layman's range of 
emotions). Why do they give us these 
feelings, and why are most of these epi- 



thets derived from analogies with our own 
bodily conditions? If we ask ourselves 
these questions, we shall little by little 
come to think of architecture not as mere 
ornamented engineering but as embodied 
emotion of a very peculiar sort, a new 
language speaking to us very directly and 
very intimately. 

If we ask whether these forms reside in 
any special part (as in the mechanics or 
the linear relations or the chiaroscuro), we 
shall find that it is not possible to assign 
one formal r61e to the structural elements 
and a different rdle to the "accessory" ele- 
ments of ornamentation and decoration. 
Effects of line may be due both to structure 
and to carving or colouring; surfaces may 
be structural units or merely coloured areas 
or both at once; ornament which could 
have been dispensed with from the point 
of view of the mechanical structure, may 
be of the utmost importance for the de- 
sign like the great S-shaped scrolls of 
Santa Maria della Salute in Venice, whose 
shape is not essential to their function. 
The form of the mullions in the window- 




tracery may be the most telling source of 
emptional effect which a stylistic period 
may have produced. Architectural his- 
torians sometimes resent a division of Eng- 
lish Gothic into geometric, flowing, and 
perpendicular periods on the basis of win- 
dow-tracery forms ; but is it really so super- 
ficial and amateurish as they would make 
out? We can date just as accurately from 
hood-mouldings and drips or technique of 
buttressing or vaulting bosses or anything 
else; but it is not a question of identifying 
the period or analysing the style, but of 
naming these according to their most ef- 
fective characteristic. And this is well 
epitomized in the tracery. 

Ornament or detail can be considered 
irrelevant only when it has no function to 
perform, and inappropriate only when its 
effect is inappropriate in the whole effect of 
the building. It may be structurally su- 
perfluous and yet formally essential. Even 
a false facade may be pardoned when it 
does not aim to deceive. If the Tuscan 
builders intended to fool the passer-by into 
thinking that there was a greater and 



higher church behind the fagade than they 
had really built there, they must have been 
simple-minded in crediting the public 
neither with curiosity nor with legs. But 
if they meant merely to make a fine front 
to their church, as one might use gold and 
leather and make a fine cover to a book, 
their fault was wholly pardonable; for 
they missed merely the refinement of taste 
which makes a fagadal design appropriate 
by deriving it out of the exigencies of the 
nave and aisles which are there terminated. 
Again, St. Peter's has a "false" dome for 
appearance's sake ! But what is that save 
an admission that the outside and inside 
surfaces of the same curvilinear form were 
felt not to be appropriate to both exterior 
and interior at once? If one shape suited 
within and a different one suited without, 
so build it ! There is no deception intended ; 
and the structural waste and additional 
cost are part of the price of the intended 
effect. Only if an equivalent effect could 
have been produced without resorting to 
such a device, would exchequer and good 
sense be right in protesting. 




So considered, ornament will be seen to 
be not an accessory or separable element, 
3ut an integral part of the builder's art. 
For this reason it will often tend away from 
pictorial representation in order that it 
may be accepted in a more intimate union 
with the other elements. 

Greek architectural ornament was largely 
drawn from forms of leaf and flower. In 
the cymatia, these leaf-forms are so con- 
ventionalised as to make their botanical 
prototypes unrecognisable. This is partly 
due to the fact that they were inheritances, 
with a long history that leads to the Orient 
and Egypt. No ordinary eye can see a 
flower in the volutes of the Ionic capital 
or a calyx in the egg-and-dart on its 
echinus; yet these are very probably the 
ultimate ancestors. But Greek architec- 
ture itself contributed to conventionalising 
its decorative forms. Though the acanthus 
leaves of the Corinthian capital are often 
carved with very considerable fidelity to 
botanical prototypes, their whole arrange- 
ment and profile prohibit any representa- 
tional illusion. And Greek architecture 




was quite content to keep its traditional 

conventionalisations rather than to dis- 

card them for new and more imitative 


At times, more representational ele- 

ments threatened to intrude. The Doric 

columns, it will be remembered, hold up 

their oppressive load and with an almost 

human analogy give, in their entasis, a 

visual suggestion that the inanimate stone 

felt the strain of bearing up architrave, 

frieze, and cornice. They are like Atlas 

holding the broad sky on his shoulders. 

The suggestion to carve these supports into 

the likeness of human figures lay therefore 

palpably close at hand. Now the basic 

impulse may have been a tendency to 

ascribe human characteristics to inanimate 

objects, to look on a column as though it 

were a living being; but the formal pur- 

pose is slightly different, for on being shown 

a fictitious strain-under-pressure in the 

column, by analogy with our own feelings 

at supporting a heavy burden, we experi- 

ence an emotional, and not merely a rea- 

soned or intellectual appreciation of the 





purpose and structural function of the col- 
umn. And when the Greeks actually 
carved giants bowed under the superin- 
cumbent load or maidens standing strong- 
hipped and stiffly straight with the burden 
on their heads they may have diminished 
rather than heightened the effect. If a 
more or less abstract form can suggest 
physical strain, we lose instead of gaining 
when we substitute for that abstract form 
a concrete symbol borrowed from the actual 
world. Probably this is so because a sym- 
bol works through the intellectual under- 
standing, while pure form works through 
immediate sensuous apprehension. 19 I 
take it to be an astonishing instance of the 
power of artistic instinct that, without ever 
very thoroughly reasoning the matter, 
architectural taste has steadily contested 
the repetition of this experiment and has 
tended to eliminate representational shapes 
from every important structural position. 

Of a similar sort is the insistence that 
vegetable forms should be conventional- 
ised, and that actual picturing of the seen 
world must be confined to structurally 



minor surfaces. But it will be objected 
that the Greeks carved and painted pic- 
tures for their metopes; that the Panath- 
enaic frieze of the Parthenon is one of the 
great achievements of sculpture; that 
Byzantine architecture revelled in mosaics; 
that Romanesque and Gothic architecture 
are full of carven beasts and statuary. 
Clearly the prejudice against mixing the 
representational with the unrepresenta- 
tional is not very thorough or far-reaching. 
The real objection, I think, is not against 
combining representational with unrepre- 
sentational art, but against allowing rep- 
resentation to obscure or destroy formal 
values. Variegations of colour (apart from 
what they depict) have an important formal 
function. Colour is not an accessory or a 
luxury to architecture. How much does 
not the little Place des Vosges in Paris owe 
merely to the red of its bricks, which charms 
us so simply and cheerily after the arid 
grey-whites of Champs-Elysian preten- 
tiousness in stone? And are we not carried 
deep into the very emotional core and 
spirit of mediaevalism merely by the colours 





of a Byzantine interior or by that lavish 
use of reds and blues and purples and greens 
and golds which so often covered the entire 
front of Gothic churches, but whose original 
presence we of to-day are always forgetting 
because we see only the " architecture," the 
building without the colour? But colour 
can even affect the structural aspect. Find 
a blank wall and cover the upper third with 
patches of- brilliant colour if you would see 
its mere physical weight lessened. Or 
consider the arrangement of horizontal 
bands of ever-lightening colour on the Pal- 
ace of the Doges. Or look again at the 
Panathenaic frieze to notice how the 
sculptural lines lead the eye along and bind 
the Parthenon about, as a ribbon holds a 
bundle together. Again, colour is a magnet 
to the eye and can bend back our heads 
and take our vision aloft quite as quickly 
and well as the guiding lines of a moulded 
pier or the branching ribs of a vaulting. 

Thus the mere occurrence of carving and 
colour need not obscure or destroy the for- 
mal values, but may itself perform a formal 
office. There remains the difficulty that this 



colour or carving is not mere colour and 
line, but makes pictures of things irrelevant 
to the architectural structure of the build- 
ing. In mediaeval times, these pictures 
were part of an attempt to wed understand- 
ing with architectural emotion. When the 
eye had imparted directly the emotions in- 
herent in the wonderfully lined and bal- 
anced, shadowed and coloured structure, 
and aroused through architectural form 
something emotionally comparable to a 
sense of striving and attainment, patience 
and perseverance, wealth and rejoicing, 
gloom and terror and triumph, it was the 
turn of the intellect to read the illustrated 
book of saints and martyrs, angels and 
devils, with their recollection of trials and 
virtuous deeds, and to wed these emotions 
of the intellect with those simpler, more 
primitive ones of the unreasoning appre- 
hension. In Greece, too, the contempla- 
tion of heroic adventures and divine legends 
was intended to satisfy the reasoning curi- 
osity and to offer its more sophisticated 
appeal as a supplement to the direct work- 
ing of architectural form. For, in both 




cases, architecture was toiling in the in- 
terest of religion; and the ultimate channel 
of its emotions lay beyond its own domain. 
We have seen that the initial barrier to 
an understanding of architectural esthetics 
lies in our own insistent assumption that 
the art is governed by practical utility, 
whereas insofar as architecture is an art 
it has almost nothing at all to do with 
practical utility. This dogma will always 
be hotly contested. "Is not architecture 
the art of building, and are not buildings 
primarily dwelling-places with walls to 
give shelter, roofs to exclude rain, windows 
to let in light, and plans arranged for 
specific human needs? and do these not 
depend absolutely on the practical laws of 
engineering and gravitational mechanics? 
How then can you say that it has nothing 
to do with utility?" But this is merely 
to say that architecture is a practical activ- 
ity first and an art only secondarily. In- 
sofar as it is an art, it has almost nothing 
to do with these practical demands. As 
an art, it is a spatial presentation of solid 
forms, aiming at no imitative pictorial 




illusion, but appealing primarily to our 
gravitational sensibilities and our powers 
of spatial apprehension and secondarily to 
certain activities of visual comparison and 
construction which we call geometric 
when they are viewed by our intellect, but 
for which we have no good name (unless it 
be Artistic Pure-Forms) when they are ap- 
prehended through our esthetic sensibility. 

"But," I hear some one say, "you must 
be a good engineer to be a good architect," 
or again, "this so-called esthetic architec- 
ture is all very well on paper; but wait till 
you try building it," or, "when a city orders 
a municipal court-house, does it want 
spatial presentations and appeals to the 
gravitational sensibilities of the tax-pay- 
ers, or does it want a good serviceable 

Most of the arts were originally slaves 
to practical ends. Poetry had to chronicle 
or instruct or placate; sculpture had to 
serve religion or superstition or human-self- 
glorification; similar tasks were assigned to 
painting; but all these, even from the 
earliest time, seem also to have existed in 




their own right, because what they did was 
directly pleasing to men . Even in the pop- 
ular mind, poetry has emancipated herself 
of any ulterior end or service; sculpture 
and painting are for the delight of what 
they show; but architecture for the most 
part still stays in bondage. Her works 
are too costly and too unwieldy to be pro- 
duced for mere contemplation; and since 
they are unrepresentational, men have sel- 
dom seen what such works might be in- 
tended for, if not just to house men and 
their chattels with perhaps a fine external 
show of wealth. Perhaps because of this 
popular blindness, though we may have 
laid a thousand stones to antiquity's one, 
we make no greater art of our architecture 
than did the ancient builders. Indeed, if 
subtlety, refinement, perfection of propor- 
tion, clarity of form, sensitiveness to out 
line and surface and the visual appearances 
of weight in equilibrium, if all these are 
criteria of artistic accomplishment, twen- 
tieth century America has not even re- 
motely rivalled the builders of ancient 




1 1 could wish that writers on esthetics 
Mrpuld be more mindful that this is so, and that 
with all artistic illusion (if it be truly artistic) 
there is fused to some degree the sensibility 
that illusion is. .different from reality. This 
applies to~alTthe imitative and presentational 
arts, whether of the theatre or the atelier; 
and it needs no very great attention to dis- 
cover that every art protects its title to 
artistry by clinging to certain contentions 
which will' obviate complete and full illusion. 
The theatre has its curtain and footlights and 
a host of conventions of acting; sculpture 
eliminates colour, or rather chromatic fidelity; 
painting must make (and particularly delights 
in making) its own world of light and air, and 
so need seldom trouble itself with the fear 
of too illusionary a content; the minor arts 
are ever conscious of the gold or silver or 
bronze or ivory or precious stone in which they 
work, and at their best are least ashamed of 
the peculiar and native qualities of their 

2 Perhaps the landscapes of Chinese paint- 
ing will best suggest the direction which Greek 
art might have taken in a less luminous and 
more misty climate. There is the same scru- 
tiny of individual forms, the same indifference 
to the merely optically correct, the same 
intellectualisation and artistic rethinking of 
objective appearance, the same extreme 
fidelity to the ideal content. 





1 There is a possible confusion here which 

ought to be avoided. The particular run of a 

line may suggest that the thing which it helps 

to represent is in motion. Lines of an appro- 

priate contour will make us sec a dog running, 
a bird flying. Such lines present motion, i. e., 

motion in the represented object. But the 

motion which is a formal function is not in the 

picture so much as in ourselves. There might 

be a greater use of formal motion in a still- 

life picture than in a representation of objects 
supposedly in very rapid motion, e. g ., a photo- 
graph of a galloping horse. 

2 This is quite in accord with later Greek 

estimates of the early painters. Compare, 

for example, the following passage from 
Dionysius of Halicarnassus (de Isaeo judic. 4) : 
" In ancient painting the scheme of colouring 

was simple and presented no variety in the 

tones; but the line was rendered with 

exquisite perfection, thus lending to these 

early works a singular grace. This purity 
of draughtsmanship was gradually lost; its 

place was taken by a learned technique, by 

bhe differentiation of light and shade, by the 

Full resources of the rich colouring to which the 

works of the later artists owe their strength." 

(Trans. Jex-Blake.) 


*. . . the mere contours will make them intelli- 

gible. Cf. Pliny (N. H., xxxv, 67-8): "... 

rendering of outline. This is the highest 




subtlety attainable in painting. Merely to 
paint a figure in relief is no doubt a great 
achievement, yet many have succeeded thus 
far. But where an artist is rarely successful 
is in finding an outline which shall express the 
contours of the figure. For the contour 
should appear to fold back, and so enclose 
the object as to give assurance of the parts 
behind, thus clearly suggesting even what it 
conceals." (Trans. Jex-Blake.) 

*Polykleitan statues are too "square" Cf. 
Pliny (N. H. t xxxiv, 56): "Quadrat a tamen 
esse ea ait Varro." 

N. H. t xxxiv, 65. 

4 In addition to this rotation of the axes 
in a horizontal plane, a vertical displacement 
of these axes occurs in more complicated poses 
such as those of sitting or crouching or 
recumbent figures. It will be found that this, 
too, is an ordered and regular process con- 
tributing in a similar manner to the same 
end. Cf. the Sleeping Satyr in Naples. 

6 There is a possible quibble here which I 
mention, if only to show that I have not over- 
looked it. The third dimension in normal 
geometric space stretches away in parallel 
planes. In visual space its planes all inter- 
sect in the eye and therefore cannot be 
parallel but must be radiate like the lines 
from a vanishing point in perspective draw- 
ing. The invisible third dimension of which 
I have been speaking in connection with the 
spatial extension of sculpture refers to this 
third or radiative dimension in visual space, 
and should be so understood. 




'Plutarch, Quaest. Conu., ii, 3, 2: 
KA.CITOS 6 7rA.a<TT?7? eiTre x a ^7ra>TaTav eirai TO 
fpyov orav ev ovu^c, 6 7n/A.os yi'vyrai. 

7 Though not in the form nor to the extent 
postulated by Mr. Jay Hambidge in his 
''Dynamic Symmetry." 

8 Galen, de Plac. Hipp, et Plat., 5. 

9 /?eA.07roiiKwv, iv, 2. 

10 By using an elaborate system of simple 
ratios. . . . Similarly Albrecht Diirer: "I 
make the rule always one-sixth of the length 
of the figure . . . then I divide the rule into 
ten equal parts and each part I call a zall, 
each zall I divide into ten and call each tenth 
a tell, each teil into three and call each third 
a triimmlein." Here are indeed "many 
numbers." The smallest unit of division is 
one 1800th part of the total height of the 
figure ! 

Pliny, N. H., xxxiv, 65. 
"Quint., xii, 10, 7. 


1 The Greeks were timid engineers. Consider 
the elaborate precautions taken with the 
central span of the Propylaea (W. B. Dins- 
moor in American Journal of Archaeology, 
1910, pp. 145 ff). By applying the modern 
engineering formulae given in Kidder's 
Architect's and Builder's Pocketbook, I find 
that these precautions were superfluous. 

8 Their columns were . . . but barely sixty 
feet. Cf. Athens, Olympieum; Miletos, 
Didymeum; Ephesos, "fifth" or Hellenistic 
temple of Artemis. 



Contrast the rich inventions of plan in 
late Roman times (Montano, Scielti di varii 
tempietti antichi, and Raccolta di tempii, etc.). 

4 1 do not accept Dorpfeld's suggestion of 
an original symmetrical plan doubled on a 
north-south axis. 

8 Plato, Republic. 

Cf. for example Dinsmopr's plate showing 
the evolution of the Ionic Order in Asia 
Minor (A.J.A., 1908, p. 4); but cf. also 
Marquand, Greek Architecture, p. 131 infra, 
" . . . the Greek love of variety, which 
makes it impossible to apply the rule me- 
chanically so as to establish an exact chrono- 
logical series." 

7 Cf. Koldewey and Puchstein, die griech- 
ischen Tempel in Unteritalien -und Sicilien. 

... they left no straight line straight. The 
material for the study of this question is to be 
found in Penrose, Principles of Athenian 
Architecture and Goodyear, Greek Refine- 

9 ... not optical corrections. Goodyear has 
abundantly proved this point in the book 
referred to in the previous note. 

"Cf. T. A. Cook, The Curves of Life; 
d'Arcy Thompson, Growth and Form; A. H. 
Church, Relation of Phyllotaxis to Mechanical 
Laws\ and the two (very popularly written) 
books by S. Cplman and C. A. Coan, 
Nature's Harmonic Unity and Proportional 

11 Mausoleum. I am here following the 
measurements and calculations of W. B. 
Dinsmoor (A.J.A., 1908, pp. 1-29) rather 




than those of Lethaby (Builder, Feb. 6, 1920, 

p. 168; Sept. 3, 1920, p. 256). 

12 The lions 1 heads . . . are so placed. 

This arrangement is certain for the Athena 

temple at Priene and quite possible for the 


13 It is difficult to be fully persuaded of the 
absence of a frieze in the Priene temple and 

the Mausoleum. The evidence against their 

presence is not in any sense conclusive. 

14 Commensurable but not measurable rela- 

tions, i. e., commensurable in terms of simple 

integral numbers and ratios, but riot measur- 

able in feet, palms, and dactyls without 

fractional parts. 

18 Marquand, Greek Architecture, pp. 144-5. 

16 Goodyear, Greek Refinements. 
17 In Mr. Donald Tovey's article Music 

in the Enc. Britt. (Xlth Ed.) by a curious 

coincidence a metaphorical comparison is 

made between Greek music and an art of two 

dimensions. The analogy is of course 

intended only in a figurative sense, but 

deserves at least a passing reference here: 

"Non-harmonic music is a world of two 

dimensions, and we must now inquire how 

man came to rise from this 'flatland' to the 

solid world of sound in which Palestrina, 

Bach, Beethoven, and Wagner live." 

18 The Baroque is a very good period to study. 
I wish to acknowledge very great indebted- 

ness to Geoffrey Scott's Architecture of 

Humanism for my general position on the 

esthetics of architecture. 

19 This seems to be inconsistent with what 





was said in the chapter on Sculpture. The 
point of the argument is the contention that 
in the case of a Caryatid we may appreciate 
strain and weight in terms of a human bodily 
experience within ourselves (precisely as in 
any other statue) but we are unable to 
ascribe them to the building. The empathy 
is sculptural, not architectural. 







University of Toronto 








Acme Library Card Pocket